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
|
|---|---|---|---|---|---|
Missing data are an unavoidable component of modern statistical genetics . Different array or sequencing technologies cover different single nucleotide polymorphisms ( SNPs ) , leading to a complicated mosaic pattern of missingness where both individual genotypes and entire SNPs are sporadically absent . Such missing data patterns cannot be ignored without introducing bias , yet cannot be inferred exclusively from nonmissing data . In genome-wide association studies , the accepted solution to missingness is to impute missing data using external reference haplotypes . The resulting probabilistic genotypes may be analyzed in the place of genotype calls . A general-purpose paradigm , called Multiple Imputation ( MI ) , is known to model uncertainty in many contexts , yet it is not widely used in association studies . Here , we undertake a systematic evaluation of existing imputed data analysis methods and MI . We characterize biases related to uncertainty in association studies , and find that bias is introduced both at the imputation level , when imputation algorithms generate inconsistent genotype probabilities , and at the association level , when analysis methods inadequately model genotype uncertainty . We find that MI performs at least as well as existing methods or in some cases much better , and provides a straightforward paradigm for adapting existing genotype association methods to uncertain data .
Genome-wide association studies ( GWAS ) have become a primary tool to elucidate the correlations between SNP genotypes and complex phenotypes in large cohorts . Association studies initially assumed the existence of genotype calls: for each sample at each assayed variant , either reference allele homozygote , heterozygote , or alternate allele homozygote . As such , methods developed for analyzing GWAS assumed the existence of such perfect-confidence genotype data . The association study design and related analysis methods have remained in force even as the field has transitioned into the sequencing era and more complete data have become available . In all situations , due to technical and financial limitations , association studies only partially assay the set of common variants in any organism . The variants included on a SNP array typically only include a small fraction of the total pool of variants present , and even sequenced variants are called incompletely and inconsistently . Furthermore , due to the low magnitude of effect of most trait-associated variants , studies prioritize sample size via multisite meta-analysis , involving genetic samples assayed on different technologies . This study design results in a complicated missingness pattern across the entire conceptual set of common variants in a sample . Yet due to shared linkage disequilibrium between different samples , this missingness can be overcome with the addition of external reference data . Genotype imputation probabilistically estimates unknown genotypes for a study sample by leveraging external reference haplotypes ascertained at a superset of SNPs [1] . Genotype calls and genotype probabilities are fundamentally different . Genotype calls are considered certain data and , under the traditional additive model , may be represented as an integer count of a reference allele present for each study individual . Direct extension of these simple models to probabilistic data is not possible , as the models are not capable of representing the additional variance component introduced by uncertainty . This is critical to understanding the traditional challenge in analyzing imputed genotype data: existing methods are not directly compatible with uncertain data , so probabilistic genotypes must be projected to a lower-dimensional approximation resembling genotypes , with a concomitant loss of information and introduction of bias . Three primary methods have been developed for the handling of uncertain data in genetic association studies . For the first method , and at one extreme , probabilities may be converted to call-like integral counts by choosing the class with the largest probability . This paradigm requires no additional modification of the analysis method for genotype calls , but all meaningful information about uncertainty is lost . For the second method , in some situations , genotype probabilities may be converted into expected counts of the reference allele: the “allelic dosage . ” This strategy attempts to maintain some of the uncertainty of the genotype estimate by allowing non-integral values: for example , when the heterozygote and reference allele homozygote classes are equiprobable and the alternate allele homozygote has 0 probability , a sample is considered to have 1 . 5 alleles . Unfortunately , this strategy is only useful when the underlying algorithm extends to nonintegral data ( for example , a generalized linear model with continuous predictor ) ; furthermore , there is no rigorous proof of the degree of bias or information loss incurred using this method . Finally , as a last method , the algorithm may be modified to directly operate on probabilities [2] . This option attempts to include all uncertainty information at the cost of additional work creating a new algorithm . In practice , this type of custom algorithm design is limited to simple GLM methods; other studies in the field creating more complex statistical models for association studies do not undertake this additional work [3–8] . In the statistics literature , Multiple Imputation ( MI; distinct from genotype imputation; [9] ) is the rigorous method of conducting analysis on probabilistic estimates of uncertain data . The details of MI are discussed in Methods . Briefly , a small number of complete datasets are randomly drawn according to the probability data . These datasets are then analyzed using any standard analysis technique that generates a normally-distributed effect estimate β and standard error estimate s . Importantly , MI is not a method of estimating hidden data , but rather a method of handling existing estimates . The performance of MI , and indeed of all imputed data analysis methods , is reliant on the quality of the underlying genotype imputation . Imputation accuracy , the agreement between predicted and true genotype , tends to vary across both imputation “quality , ” as estimated by most imputation software , and minor allele frequency . The extent of this variable performance has not , to our knowledge , been rigorously assessed . In this study , we seek to rigorously evaluate this variable genotype imputation performance . We show a significant deviation between genotype probabilities generated by imputation , and empirical probabilities estimated at the same sites . This failure of probability consistency is an important confounding effect in imputed data analysis . We show that Multiple Imputation matches or improves upon performance of existing imputed data analysis regimes by better prioritizing true positive associations , while additionally being straightforwardly extensible to future analysis algorithms .
We compared the allelic dosage ( Methods ) to the “true” genotype count based on masked genotype data . Fig 1 shows the fraction error between allelic dosage from imputation and masked genotype , stratified by genotype class , allele frequency , and reported imputation quality metric ( hereafter called “r2” ) . We observe that a single quality metric r2 masks significant deviations in mean quality between different genotype classes and different allele frequencies . For imputed SNPs reported to be of high quality ( r2 > 0 . 9 , left panel; top 40% of GWAS-used SNPs , S1 and S2 Figs ) , and with sufficiently high minor allele frequency , imputation is indeed well-behaved: variants with minor allele frequency above 0 . 3 have less than 3% discordance with similar performance across genotype classes . However , SNPs of lesser imputation quality ( 0 . 8 ≤ r2 ≤ 0 . 9 , right panel; approximately 18% of GWAS-used SNPs , S1 and S2 Figs ) or low minor allele frequency are inconsistently imputed . Minor allele homozygotes and heterozygotes in particular are subject to highly inflated error rates . While differences across the SNP strata are observed with both IMPUTE2 and minimac3 imputations , the magnitudes observed are distinct ( S3 Fig ) . Equivalent results are observed when evaluating performance by fraction of best-guess genotypes from imputation not matching masked genotypes ( S4 Fig ) . We next examined the imputation probabilities themselves , to evaluate whether probabilities generated by imputation software correspond to the empirical probability of observing a genotype at a particular site . Results for this comparison for IMPUTE2 probabilities are shown in Fig 2 , across strata of reported quality and predicted call probability . The empirical accuracy significantly deviates from the predicted , and much more so with decreasing r2; S5 Fig shows similar plots comparing the effect of decreasing minor allele frequency on this distortion , and show weaker but significant changes with decreasing frequency . Of note , the heterozygote genotype class behaves in a distinct but complementary fashion relative to the two homozygote classes . These results are distinguishable between imputation programs: the effect is much stronger in the IMPUTE2 imputation . The most substantial difference between the two programs is IMPUTE2’s use of sequential imputation windows to improve performance through parallelization , with potential accuracy tradeoffs , yet we have observed no differences caused by modifying this parameter . We note that differences in imputation performance under different conditions have been observed extensively and unpredictably ( see , among many , [10–14] ) . Our observations are consistent with intermittent observations of MACH-family algorithms nominally outperforming IMPUTE-family algorithms in some cases , yet the precise reason ( s ) for these differences between similar software has never , to our knowledge , been demonstrated . We next sought to evaluate whether the ability to prioritize verified trait-associated SNPs in an association study ranking was detectably different using MI or other existing algorithms . We considered 73 replicated loci from a large ( N = 339224 individuals ) GWAS for BMI [15] . As expected with our modest sample size of 2802 , we have little power to detect the majority of these variants at genome-wide significance 5 ⋅ 10−8 . Nevertheless , if the variants are associated with the trait at all , one expects the variants to be relatively better ranked in the final list of variants than variants chosen at random from the study . We conducted a BMI genome-wide association study in Health ABC , as discussed in Methods . Using these SNP association results , we computed the rank percentile of each published variant , comparing these results for two existing imputed data analysis algorithms and MI . We evaluated the receiver operating characteristic curves for PLINK dosage , SNPTEST score , and MI ( Fig 3 ) . MI significantly outperforms all other methods ( one-tailed DeLong test p < 2 ⋅ 10−16 ) . We repeat this analysis with a height GWAS in NFBC66 ( Methods ) , and find a similar significant improvement in signal detection by MI compared to other methods ( one-tailed DeLong test p < 0 . 0003764 ) . We conclude from this analysis that significant information loss may occur when uncertainty is incompletely handled in imputed data analysis . There is no evidence in this comparison that Multiple Imputation is inferior to other methods . Improved signal detection by MI may be attributed to several causes . In observing the additional variance component in the multiple imputation model , s B 2 ( see Methods ) , we note that variability introduced by genotypic uncertainty should result in decreased rankings for variants regardless of trait being analyzed . Uncertainty in probabilities also differs with variant-wide imputation quality r2 , implying that different association tests may lead to different expected distributions of variants in a ranked association study across the quality spectrum , whether variants are trait-associated or not . We sought to evaluate the null distribution of variant ranks across imputation quality . We calculated the percentile rank of each variant in the BMI GWAS and sorted the variants into bins across imputation quality . Results from this investigation are shown in Fig 4 . All tested methods have a statistically significant correlation between imputation quality and average rank ( all Pearson correlation test p < 2 ⋅ 10−16; comparable results for nonparametric tests ) . PLINK and SNPTEST have indistinguishable magnitudes of effect ( shared effect size -0 . 017 , indistinguishable with p = 0 . 44 ) . Multiple Imputation shows a significantly stronger correlation than the other tests ( effect size -0 . 351 , greater than other tests with p = 3 . 4 ⋅ 10−7 ) . MI produces test statistics that much more significantly incorporate uncertainty , through in particular the between-draw variance component . Published trait-associated variants tend to impute with higher quality than random variants from a dataset due to selection bias , leading to the possibility that this global incorporation of quality by MI is leading to better performance in a manner unrelated to the trait itself . To control for trait-unrelated differences between published variants and the global SNP distribution , we computed empirical matched null sets for every published variant by drawing 1000 SNPs matched to this variant with parameters r2 ± 0 . 01 and frequency ±0 . 01 of the true variant . Here we are drawing variants from the association study itself; this null allows us to detect effects specifically affecting the significant tail of the distribution over the bulk of variants . For most traits , these variants will overwhelmingly be unassociated with the outcome , and thus this null will correspond to a true null of no genetic association; in the case that this condition is untrue , this null will lead to relatively conservative assumptions about the added performance of MI . We then regenerated ROC curves for each analysis , this time controlling for the null rank of SNPs matched to the published variant list based on these parameters . These results approximately correspond to trait-specific enrichment effects caused by the various analysis methods . We find that adjustment for trait-secular shifts in variant quality affects BMI and height differently . For BMI , MI continues to outperform PLINK and SNPTEST ( one-tailed DeLong p < 1 . 177 ⋅ 10−10 ) . For height , null adjustment reverses the previously-observed trends , such that MI tends to underperform PLINK and SNPTEST ( one-tailed DeLong p < 0 . 0025 ) . We note that for this trait in particular , the null derived from drawing from the association study itself may be conservative given the broad genetic basis for human height [16] . To underscore the benefits of MI not simply on regression with existing uncertainty handling methoods , but additionally on more complex algorithms more difficult to directly adapt , we applied MI to EMMAX [4] . We ran EMMAX on both the Health ABC BMI GWAS and the NFBC66 height GWAS , with default parameter settings and the IBS-based kinship matrices . We compared EMMAX with thresholded genotype data and MI on imputed probabilities . The published version of EMMAX only accepts integral count genotypes , thus no other comparisons were included in this test . We detect significant improvement in relative percentile increase for BMI-associated variants that is abrograted by null adjustment ( one-sided DeLong test p = 6 . 663 ⋅ 10−12 and p = 0 . 09933 respectively ) . We find weak evidence for improved overall efficiency for known height variants ( one-sided DeLong test p = 0 . 01994 ) , which under null adjustment becomes a significant underperformance of MI relative to rounded genotype control ( one-sided DeLong test p = 2 . 242 ⋅ 10−7 ) ; again , we note that this null may be overly conservative in this trait context [16] . Evaluation of the null distribution of variants for this experiment , analogous to the analysis in Fig 4 , shows that the MI-mediated rank shift is much smaller across both the frequency and variant quality spectrum . We propose that the error model used in EMMAX is partially compensating for the additional uncertainty variance component redundantly modeled by MI .
In this paper , we analyze the application of Multiple Imputation in the particular context of genotype imputation . We find that representative existing analysis methods tend to increase noise in association testing by incompletely modeling the uncertainty of genotype estimates across the entire set of imputed variants . This results not just in lower quality variants , for which information is limited , receiving inappropriately high ranking , but variants for which underpowered but significant trait association is present becoming lost in statistical noise introduced by imputation . We furthermore detect imputation program-specific inconsistencies in posterior genotype probabilities that differently affect the three genotype classes . Overall , we characterize the complexity of imputation probabilities and show that MI can improve association testing with uncertain data at the cost of increased post-imputation computational time . This study particularly elaborates on the interpretation of a p-value from a probabilistic association study . Based on the results in Fig 4 , we see that existing analysis tends to rank variants uniformly across r2: the p-value weakly incorporates imputation quality , enough so that the tests are not correctly calibrated but insufficiently so to actually correct for differential uncertainty . This poor calibration is verifiable with a true null: observing the null distribution of variants in our study regressed against random standard normal traits , we observe a comparable correlation between variant imputation quality and SNP ranking for SNPTEST ( p = 0 . 001953 ) . Dosage analysis does not , in that context , recapitulate the correlation effect observed in our data , suggesting that the trend is specific to our trait-associated null . This poor calibration is potentially shared by many analysis methods in the field: following the simulation work of Acar and Sun [17] , we have found low magnitude but significant miscalibration when operating on probabilistic genetic data and null traits for linear regression and a Kruskal-Wallis test on best-guess variants; dosage analysis; and their generalized Kruskal-Wallis test that directly handles uncertainty ( all Wilcoxon rank sum p < 2 ⋅ 10−16 ) . There is also a significant correlation between uncertainty for the Kruskal-Wallis based tests in this simulation context ( ANOVA p < 2 ⋅ 10−16 ) . Association p-values are intrinsically affected by genotype uncertainty , altering the expected null distribution from uniform random , but this is simply an uncharacterized bias introduced by incompletely handling uncertainty , not a formal characteristic with proven properties . We suggest that this behavior is undesirable: if a SNP estimate is uncertain , then the behavior of an ideal inference should be to proportionally downweight the ranking of that SNP in that particular study in a predictable fashion . With Multiple Imputation , this uncertainty-induced variability is correctly apportioned to the between-draw variance component s B 2 , resulting in unmodified effect estimates that are still suitable for cross-cohort meta-analysis . In the case of our example BMI association study , the positive control variants used were all imputed with high confidence , yet MI still showed a trait-specific rank improvement in these SNPs , due to the selective downranking of low confidence variants with no or lesser phenotype association . In short , we see little reason to continue using dosage or score test methods for handling imputation probabilities , when a straightforward alternative with superior statistical properties exists . The results in Figs 1 and 2 broadly characterize imputation performance across variant quality , frequency , and genotype class . Although parts of these results are hinted at in various imputation papers ( in particular , [18] , but also [2 , 12 , 19 , 20] ) , we have not before encountered a presentation of the severity of quality degradation across the full spectrum of frequencies . We urge analysts in all contexts to avoid genotype thresholding , in particular using IMPUTE2 probabilities: with decreasing frequency and quality , the likelihood of rounding to an incorrect genotype grows extremely high . We also observe strong per-genotype class differences in performance , most severly affecting heterozygotes . This is not a surprising result , given the integral nature of computational prephasing in genotype imputation . In the modern genotype imputation protocol , prephasing is a one-time burdensome computation that is not repeated , whereas the imputation step itself is comparatively rapid and repeated many times as new reference panels become available . A useful future analysis would investigate the specific impact of differential phasing quality on imputation probability bias , and our results emphasize that prephasing should be reconducted as advancements in the field yield higher quality phasing solutions . We have observed in the literature a tendency to ignore uncertainty in genotype data when developing new algorithms . This elision is understandable in the sense that complete-case analysis is typically substantially more straightforward to implement , and as a first approximation , uncertain data may be converted seemingly straightforwardly to genotype “calls” via techniques such as rounding . Yet in the case study of simple regression , we see that this approximation bears with it a cost in loss of statistical power to detect trait-associated variants . One of the great benefits of MI is its simple application to complete-case methods , as each round of MI generates complete case data . In the case of standard regression , this is a straightforward benefit; in the case of a method such as EMMAX , one must balance the desirable avoidance of rounded probabilities with potential complications of the standard MI variance component model . One of the strongest justifications for the widespread use of imputation is the facilitation of multisite multicohort meta-analysis , in which summary statistics from separate association studies are combined to increase statistical power . The benefits of integrating imputation quality into inference are magnified in this context . Under the current regime , either {β , s2} or {N , p} pairs are the exclusive data provided to meta-analysis tools [21] , leading to a downstream analysis that treats different estimates of variants with different imputation qualities as estimates identical to one another in expectation ascertained with perfect confidence . With nested model multiple imputation ( [22] , Methods ) , the improved model described in this paper may be extended to meta-analysis . A nested model MI meta-analysis will explicitly compensate for variable imputation quality across contributing studies with inflated variance components corresponding to noise in the mean effect estimate or inflated standard error from contributing studies ( i . e . , imprecision in effect estimate introduced either by variable imputation quality , fluctuations in allele frequency , or differential effect due to LD changes in different studies ) . The cost of this extended meta-analysis regime is limited . In addition to the standard {β , s2} pairs currently used to combine analysis , the within- and between-draw variance components must also be submitted . These data require additional disk space but are relatively trivial to provide . Crucially , the summary statistics from each individual draw are not required for this meta-analysis , such that the growth in memory requirements for meta-analysis does not scale with the number of draws conducted . We note that by explicitly handling variable imputation quality in different studies , this meta-analysis regime introduces potential sources of heterogeneity in the final meta-analysis result . Studies with aberrantly high uncertainty may strongly influence the resulting meta-analyzed association statistic . Note that this source of heterogeneity already exists , but is not rigorously modeled and currently must be addressed by ad hoc filtering and quality control . We propose that heterogeneity from differential imputation quality may be quantified by a heterogeneity test analogous to the effect estimate heterogeneity I2 metric in METAL . This test would quantify heterogeneity specifically in the between-test variance component s B 2 from contributing studies , which captures the noise between individual MI draws introduced by meaningfully uncertain data . This test would enable standardized detection of cases in which low imputation quality may require custom secondary genotyping for validation . In the statistics literature , starting with [9] and moving forward , the recommended number of MI draws has varied widely . The original recommendation was that oftentimes 3–5 draws were sufficient to retain most of the accuracy while minimizing computational burden . More recent publications ( i . e . [23] ) have suggested that the original estimates of draws were insufficiently stringent . In this paper , we have used an a priori setting of 10 draws after testing various draw numbers’ effects on quality of MI output , and balancing this impact with the added burden of multiple additional MI rounds . We find that this draw count surpasses the point at which MI effect estimates tend to converge ( S6 Fig ) , but may in the case of particularly poorly imputed variants lead to suboptimal estimates of between-draw variance ( S7 Fig ) . A useful target for future work would be the integration of dynamic computation of the number of draws for convergence for individual variants , which in this study is complicated by the manner in which our software interfaces externally , rather than reimplements internally , with existing analysis tools to maximize compatibility . The existence of program-specific differences in the consistency of imputation probabilities is an intriguing result that raises meaningful questions for the field . Though each program has its own algorithmic features and drawbacks , the practical choice between imputation software often reduces to the banal: ease of use , runtime and memory requirements . These features are important for imputation study designs that may take weeks to months to complete . Yet here we show a more substantive trend of probability bias that follows a distribution specific to particular programs . The precise cause of these distortions is not clear . Further investigation may be warranted to determine a method of adjusting the native probabilities generated from imputation using empirical distributions based on masked comparison data . We interpret the potential performance improvement of MI over other methods as a call to reevaluate the use of thresholded genotype calls in other contexts . In particular , as the field continues transitioning to sequence-derived variant data , the impact of thresholded certainty in sequencing data analysis cannot be overlooked . We note that standard methods for rare variant analysis , including burden testing , adapt to the high uncertainty in low frequency variant calls by using strict quality control and relying on bulk information to resist the noise introduced by false positives . Yet interpreting the probabilistic output of sequencing technology as a mechanism for completing data in a NMAR setting , Multiple Imputation straightforwardly provides a consistent solution for even complex statistical analyses . In the case of statistical tests that analyze linked variants together , per-site marginal information is no longer sufficient , as each MI draw must be taken from the joint distribution of all tested sites . Yet with sequencing read data , this kind of analysis is not impossible: joint information can be directly estimated from shared reads , creating a computationally challenging and yet feasible method of rigorously handling genetic uncertainty without even linkage assumptions . The creation of a standard module for interfacing with sequencing read data and dynamically generating such joint probability information would be a significant contribution to the field . Similarly , the independent draws conducted in multiple imputation implicitly assume independence of samples . In the case of pedigrees or cryptic relatedness of samples , the sampling problem becomes much more challenging . The imputation software analyzed in this study ( IMPUTE2 , minimac3 ) is designed to handle unrelated population-based samples . The marginal probabilities generated by the algorithms are not reflective of explicit joint distribution between related samples and thus are not directly compatible with correlated genotype draws without additional modeling . One could synthesize probabilities that scale with both the imputed probabilities and the Mendelian transmission rules , for example P ( child | data , parents ) = P ( child | data ) P ( child | parents ) ∑ i = 0 2 P ( child = i | data ) P ( child = i | parents ) Unfortunately , such an approximation of the joint distribution would not be equivalent to actually modeling the combined structure of the data in the imputation software itself: for example , when the parents and children are modeled separately , certain genotypes will be given nonzero probabilities that would be rendered impossible by Mendelian transmission . In the case of BEAGLE [11] , duos and trios are specially handled according to externally specified pedigrees . The resulting genotype probabilities then are reasonably considered to be conditional probabilities: for example P ( child|data , parents ) . In this case , the appropriate probabilistic relationships should hold: for example , P ( child = a a ) = P ( mother = a a ) + P ( mother = A a ) 2 · P ( father = a a ) + P ( father = A a ) 2 and thus consistent drawing could be conducted in a trio by first drawing the child , then drawing the parent probabilities conditional on the result of the child’s draw . Ultimately , the suitability of MI to a given application is restricted to situations where the probability-generating method is itself appropriately suited . Recent work ( [24–26] ) has produced algorithms ( MIX , DISTMIX , ImpG-Summary ) capable of imputing association statistics from summary data , without the need for individual-level genotype information . These methods offer substantial time savings relative to genotype imputation , at the cost of reduced overall quality of estimates relative to existing HMM methods . The imputed test statistic at a particular site is the mean of a multidimensional Gaussian distribution based on neighboring test statistics and linkage disequilibrium data: this intuitively corresponds to an MLE dosage estimate from genotype imputation probabilities . It is likely the method thus suffers from analogous disadvantages to those of incomplete uncertainty handling in genotype imputation . One could imagine replacing a single mean imputed test statistic with instead a set of random variates drawn from the underlying Gaussian; these draws would act as drawn genotypes in the analysis in this study . Although multiple test statistics per variant would prove less useful for analysts of individual studies , they would be more completely reflective of the uncertain nature of these estimates . Furthermore , using sample size weighting , one could generate an MI regime in multicohort meta-analysis in , for example , METAL , in which each set of drawn test statistics is used to generate a separate meta-analysis dataset , and the resulting sets are combined using MI . Such a regime would require further investigation , and most likely would require a substantial number of iterations if many contributing cohorts used uncertain input data . The use of MI is not completely foreign to the field of statistical genetics . We note , however , that its use is very limited , and has not been extensively compared to other , prevalent methods of handling genotype probabilities . We evaluate MI in comparison to existing methods and show MI performance is typically comparable to existing methods , and in certain contexts significantly outperforms other existing algorithms . Furthermore , we emphasize the ease with which MI is extended to , conceptually , all existing and future genotype call analysis methods with little additional effort . We foresee MI as a simple and effective component of all probabilistic genotype analysis .
For this project , we applied for access to and downloaded two SNP array and phenotype datasets from the Database of Genotypes and Phenotypes ( dbGaP ) [27] . The Whole Genome Association Study of Visceral Adiposity in the Health Aging and Body Composition ( Health ABC ) Study , dbGaP accession phs000169 . v1 . p1 , contains 2802 individuals genotyped on the Illumina Human1M-Duo SNP array . STAMPEED: Northern Finland Birth Cohort 1966 ( NFBC1966 ) , phs000276 . v2 . p1 , contains 5415 individuals assayed on the Illumina HumanCNV370 array . Both datasets were requested and approved under project 7955; they are available from the dedicated General Research Use collection and do not require IRB approval . The phenotype data released under this collection are quite limited . For this study we limited analysis to simple anthropometric traits . For Health ABC , we conducted a BMI association study with BMI ( μ = 27 . 4 , σ = 4 . 77 ) determined by age ( μ = 73 . 6 , σ = 2 . 87 ) and sex ( 51 . 2% female ) . For NFBC66 , we conducted a height association study with height in centimeters stratified by sex ( μW = 153 . 2 , σW = 68 . 9; μM = 286 . 1 , σM = 66 . 4; 52% female ) ; age was not included for this single-year birth cohort . Datasets on dbGaP have already been subjected to a round of cleaning by their depositors . Nevertheless , for thoroughness we cleaned the SNP array data using a standard QC protocol . Briefly , variants with minor allele frequency less than 1% , Hardy-Weinberg Equilibrium p-value and either per-individual or per-SNP missingness greater than 5% were removed . Cryptic relatedness was estimated using genome-wide IBS estimation in PLINK [28] . A large cluster of approximate first-cousins was detected ( IBS π ^ ∼ 0 . 125 ) . For the purposes of this analysis , whether this is indicative of unreported pedigree structure or technical artifacts in genotype collection is irrelevant , as we are not undertaking novel variant discovery , rather conducting comparisons relative to control data . The remaining SNPs were pruned to an independent subset of SNPs using PLINK--indep with default parameters , and these variants along with the maximal independent subset of individuals were used for unrooted principal component analysis in EIGENSOFT [6] . Standard population stratification along geographical axes is observed , confounding novel variant discovery but not effecting within-sample comparisons . To prepare for genotype imputation , SNPs with complementary variant alleles were removed from the dataset , and positions and SNP rsIDs were updated to those of the 1000 Genomes reference panel we used . Complementary allele variants are challenging to reconcile with datasets of potentially different strand alignments . This is true when handling external datasets subject to unknown prior manipulation , but in particular in the case of modern Illumina arrays , which are only annotated with challenging “TOP/BOT” annotations instead of reference strand calls . Following modern genotype imputation guidelines [12] , we first prephased our data using SHAPEITv2 [29] with the recommended parameters . Phased haplotype data were then probabilistically imputed to the Version 3 1000 Genomes Phase 1 Integrated global reference haplotypes [30] using IMPUTE2 [19] . For comparison purposes , to establish whether effects observed were specific to the software in use , in parallel we phased the genotype data using MACH [18] and imputed the resulting phased haplotypes to the same reference panel using minimac3 [20] . Using the global reference data , a large proportion of the approximately 40 million variants in the reference dataset are expected to not segregate in the study samples and impute very poorly; thus , before downstream analysis , variants with program-specific quality metric less than 0 . 4 were removed entirely . No per-genotype filtering was conducted , to avoid reintroducing NMAR bias . For standard association models under generalized linear models ( in this case linear regression ) , various methods currently exist for analyzing probabilistic genotypes . We selected two widely-used methods of analysis for the purposes of this study . The genetics software PLINK , as of version 1 . 07 , has an “allelic dosage” method of genotype imputation , in which the additive predictor dosage = 2P ( aa ) + P ( Aa ) is included in a generalized linear model ( note that this value is not equivalent to a perfect confidence genotype , as it is permitted to be a decimal number between 0 and 2 ) . This method projects the two parameter posterior probability into a single dimension , thus losing information in many cases: for example , this would consider the posterior probabilities {0 , 1 , 0} and { 1 3 , 1 3 , 1 3 } to be equivalent . The IMPUTE2 software used for imputation in this study has an accompanying analysis software package , called SNPTEST [2] . This software comes with a custom score test for explicitly handing genotype probabilities from imputation . The software has changed substantially since initial release; the best documentation available for the probability-handling methods is at https://mathgen . stats . ox . ac . uk/genetics_software/snptest/snptest . v2 . pdf . For this investigation , we solely use SNPTEST’s frequentist methods , which can explicitly handle uncertainty . Genotype imputation is a discipline-specific solution to a general statistical problem called “informative missingness” [9] . Consider a conceptual data matrix Y containing all phenotype and covariate data for a study . Classical statistical analysis assumes Y is complete , containing no null entries , and that all entries are known with perfect confidence . Study designs with null entries can be compelled into this format by removing all null datapoints before statistics are performed , resulting in a so-called “complete-case analysis . ” The effect of deviations from these assumptions vary depending on the characteristics of the missingness itself . Now assume that Y is the conceptual matrix containing all true datapoints with perfect confidence . Missingness observed in realistic studies is encapsulated by a second matrix , M , where each entry Mij is 1 if Yij is missing in the true study , and 0 otherwise . Using this framework , missingness can be partitioned into three general classes . If the distribution of M is independent of Y , the missingness is called “missing completely at random” ( MCAR ) . In this situation , corresponding to the classical model , all missingness can be safely ignored in downstream analysis , with a potential loss in statistical power but no introduction of bias . If instead the distribution of M is dependent on the data matrix Y , missingness can be classified in two separate cases . If the dependency can be reduced to simply the observed subset of Yobs , where Y = Yobs ∪ Ymis , then the missingness is , somewhat misleadingly , termed “missing at random” ( MAR ) . Data that are MAR cannot be ignored while safely avoiding the introduction of bias . However , due to the restriction on Yobs , the missing values can be probabilistically imputed from Yobs and added into downstream statistical analysis . In the worst case , the distribution of M is irreducibly dependent on Y; such missingness is called “not missing at random” ( NMAR ) . NMAR data cannot be removed without potentially introducing bias , and furthermore cannot be predicted solely from the observed data Yobs . The case of genetic data collection is invariably NMAR , as missingness created by collection technologies such as SNP arrays and sequencers exhibit different performance at different underlying genotypes , and the selection of variants for SNP arrays is itself biased by numerous factors including but not limited to predominant ancestry in early variation projects such as HapMap , variant location in the genome and neighboring sequence content , and allele class at the site of interest . Genotype imputation is an attempt to project the NMAR missingness of genetic data collection into an MAR condition . Due to linkage disequilibrium ( LD ) , the nonrandom segregation of neighboring variants over a limited number of generations , one can add externally collected , ancestrally related haplotypes to the data ( the Y matrix ) . The resulting partition of this matrix Yobs now ideally contains sufficient information to probabilistically estimate missing genotype data in the original dataset at both typed and untyped variants . The framework for Multiple Imputation is established in [9] . Briefly , the method assumes that missing datapoints have been probabilistically estimated using some external method . From these probabilities , an arbitrary d complete datasets are drawn from the probability distributions . In the case of standard single-variant analysis , conducting these draws is straightforward as linkage disequilibrium can be ignored . Each draw is independently subjected to the desired statistical test . This results in d sets of { β i , s i 2 } effect and standard at each variant . The Multiple Imputation consensus test statistic is computed from the following values: βMI=1d∑i=1dβisW2=1d∑i=1dsisB2=1d−1∑i=1d ( βi−βMI ) 2sM I2=SW2+ ( 1+1d ) sB2 Here , βMI is the consensus effect estimate; s W 2 is the within-draw sample variance; s B 2 is the between-draw sample variance; and s M I 2 is the total sample variance . The test statistic is the ratio of the test statistic and sample standard error , β M I s M I 2 , which is distributed as T with ( d−1 ) ( 1+dsW2 ( d+1 ) sB2 ) 2 degrees of freedom . The resulting probability may be interpreted as a posterior probability incorporating both the evidence for association in the study and the actual reliability of the genetic data . Software implementing this Multiple Imputation regime ( in beta ) may be found at https://github . com/cpalmer718/statgen-mi . This package features modularized , extensible interfacing with existing analysis software , and bsub/qsub integration . In total , this implementation of MI requires d times as long to run , and d times as much disk space , as a single analysis run , though running in tranches per chromosome on a cluster can reduce the maximum memory and effective time use by removal of intermediate files and quasiparallelization . With Multiple Imputation , a simple regime for seamlessly correcting for different proportions of uncertainty in the contributing analyses is available . Extending the logic used when combining d individual MI draws , the following meta-analysis regime applies [22] , for M contributing cohorts to a multisite meta-analysis: β^=1M∑i=1MβMI , is^W2=1M∑i=1MsMI , W2s^B2=1M∑i=1MsMI , B2s^meta2=1M−1∑i=iM ( βMI , i−β^ ) 2 The new variance component s ^ meta 2 is the between-site variance of estimated test statistics . Assuming a balanced study design in which each site runs the same number of MI rounds , the total variance of this nested model multiple imputation is s ^ 2 = s ^ W 2 + ( 1 + 1 d ) s ^ B 2 + ( 1 - 1 M ) s ^ meta 2 . The ratio of β ^ to s ^ 2 is distributed approximately T with degrees of freedom 1M ( d−1 ) ( ( 1−1d ) s^W2s^2 ) 2+1M−1 ( ( 1+1M ) smeta2s^2 ) 2 . In the case of imbalances in the number of draws conducted in each contributing cohort , more complex expressions might be derived , or alternatively a conservative estimate of min ( dm ) may be used for the weighting factor in the total variance and degrees of freedom .
|
Genetic research has been focused at analysis of datapoints that are assumed to be deterministically known . However , the majority of current , high throughput data is only probabilistically known , and proper methods for handing such uncertain genotypes are limited . Here , we build on existing theory from the field of statistics to introduce a general framework for handling probabilistic genotype data obtained through genotype imputation . This framework , called Multiple Imputation , matches or improves upon existing methods for handling uncertainty in basic analysis of genetic association . As opposed to such methods , our work furthermore extends to more advanced analysis , such as mixed-effects models , with no additional complication . Importantly , it generates posterior probabilities of association that are intrinsically weighted by the certainty of the underlying data , a feature unmatched by other existing methods . Multiple Imputation is also fully compatible with meta-analysis . Finally , our analysis of probabilistic genotype data brings into focus the accuracy and unreliability of imputation’s estimated probabilities . Taken together , these results substantially increase the utility of imputed genotypes in statistical genetics , and may have strong implications for analysis of sequencing data moving forward .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"applied",
"mathematics",
"variant",
"genotypes",
"population",
"genetics",
"alleles",
"genetic",
"mapping",
"simulation",
"and",
"modeling",
"algorithms",
"probability",
"distribution",
"mathematics",
"statistics",
"(mathematics)",
"test",
"statistics",
"genome",
"analysis",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"mathematical",
"and",
"statistical",
"techniques",
"statistical",
"methods",
"genetic",
"loci",
"probability",
"theory",
"haplotypes",
"heredity",
"meta-analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"evolutionary",
"biology",
"computational",
"biology",
"human",
"genetics"
] |
2016
|
Bias Characterization in Probabilistic Genotype Data and Improved Signal Detection with Multiple Imputation
|
Characterization of the evolutionary constraints acting on cis-regulatory sequences is crucial to comparative genomics and provides key insights on the evolution of organismal diversity . We study the relationships among orthologous cis-regulatory modules ( CRMs ) in 12 Drosophila species , especially with respect to the evolution of transcription factor binding sites , and report statistical evidence in favor of key evolutionary hypotheses . Binding sites are found to have position-specific substitution rates . However , the selective forces at different positions of a site do not act independently , and the evidence suggests that constraints on sites are often based on their exact binding affinities . Binding site loss is seen to conform to a molecular clock hypothesis . The rate of site loss is transcription factor–specific and depends on the strength of binding and , in some cases , the presence of other binding sites in close proximity . Our analysis is based on a novel computational method for aligning orthologous CRMs on a tree , which rigorously accounts for alignment uncertainties and exploits binding site predictions through a unified probabilistic framework . Finally , we report weak purifying selection on short deletions , providing important clues about overall spatial constraints on CRMs . Our results present a complex picture of regulatory sequence evolution , with substantial plasticity that depends on a number of factors . The insights gained in this study will help us to understand the combinatorial control of gene regulation and how it evolves . They will pave the way for theoretical models that are cognizant of the important determinants of regulatory sequence evolution and will be critical in genome-wide identification of non-coding sequences under purifying or positive selection .
Gene regulation is well recognized as a major determinant of how an organism functions [1] , and is also gaining recognition as an important evolutionary substrate [2] , [3] . Transcription control is one of the most common forms of gene regulation , and is known to be implemented through regulatory sequences often in the neighborhood of genes . Binding of transcription factors ( TFs ) to certain positions within regulatory sequences enhances or inhibits transcription and these bound sequences are called transcription factor binding sites ( TFBSs ) . In the case that a gene has to be combinatorially regulated by multiple transcription factors , the cognate TFBSs of those regulating factors tend to be clustered together in ∼1 Kbp-length sequences called “cis-regulatory modules” ( CRMs ) , or simply “modules” [4] . Despite significant recent efforts [5]–[8] , we lack a good understanding of the organizational principles of CRMs , e . g . , the requirements on strengths and arrangements of binding sites within a particular CRM . Inter-species comparison of modules provides a major opportunity to improve our understanding of such principles: ( i ) Evolution of CRM sequences is constrained by functional requirements , so the study of CRM evolution should allow us to infer which underlying features are more important , and to what extent . ( ii ) One may hope to find certain evolutionary signatures of CRM sequences through careful inter-species analysis [9] , greatly facilitating the identification of yet unknown CRMs . ( iii ) The study of CRM evolution will also enable us to better understand the path “from DNA to diversity” [10] . Transcription factor binding sites are commonly predicted based on the assumption of their evolutionary conservation [11] . However , the exact nature of their conservation presents a complex picture . The study by Moses et al . [12] in yeast revealed that the rates of change of nucleotides of a TFBS depend on the binding profile of that TF–the positions of more specific protein-DNA binding permit lower rate of change . It should therefore be possible to leverage the position-specific substitution pattern to better predict TFBSs , as was done in [13] . This pattern has also been reported in bacteria [14] and vertebrates [15] , but not in Drosophila . Given that this evolutionary pattern has already been assumed in practical analysis [16] , it seems worthwhile to verify it in Drosophila . Moses et al . [13] further assumed that evolution of nucleotides at different positions are independent , and existing models of binding site evolution [17] , [18] rely on this assumption; however , its validity is not obvious , given that a binding site typically functions as a unit . Empirical evidence either for or against this assumption has been lacking , except for a study in bacterial evolution [19] ( where the evidence was against it ) . There is thus a clear need to test existing and new models of binding site evolution on the multi-species data from different phyla . Even the most fundamental assumption of regulatory comparative genomics , that binding sites are evolutionarily conserved , has been challenged–Emberly et al . [20] found that binding sites are not substantially more conserved than their adjacent sequences in Drosophila; also , TFBSs are often found to have an unexpected amount of flux ( gain or loss ) in known CRM sequences [21]–[23] and in TF-bound regions in in vivo binding assays [24] , [25] . It has been suggested that this flux is in part due to expression changes in the genes controlled by these sequences [24] , and in part due to weak selection on individual sites even if the expression pattern of the target gene is conserved [26] . However , quantitative estimation of the strength of selection on binding sites has rarely been made , and requires extensive data on sets of orthologous binding sites . Moreover , the question of what leads to the observed levels of TFBS loss and gain is far from being resolved . For example , are the sites with higher binding affinities more likely to be conserved in evolution ? How does the local context , i . e . , the presence of other sites in the neighborhood , affect the probability of loss of a site ? Does the loss probability correlate with overall selective pressure ( substitution rate ) of the CRM ? Cameron et al . [27] showed that insertions or deletions ( “indels” ) may be a powerful predictor of CRM sequences in sea urchin , as long indels were suppressed inside CRMs relative to their neighboring sequences . Lunter et al . [28] speculated that such a selection pattern may be particularly relevant to CRMs , as the “fitness” of these sequences may be sensitive to the length of the sequences between adjacent TFBSs , but not their exact nucleotide composition . In several earlier studies involving a number of well-studied CRMs in Drosophila , such a pattern has not been fully observed [23] , [29] . So the following question remains: is indel-purifying selection in regulatory sequences a general evolutionary force , common to different organisms ? The answer will affect our understanding of CRM organization; e . g . , how tolerant a CRM sequence is to the change of spacing between TFBSs . Earlier attempts to characterize the evolutionary patterns of regulatory sequences used a few well-studied CRM sequences . These studies were limited in their scope [21] , [23] , [29] . The availability of 12 Drosophila species [30] and a large collection of experimentally verified Drosophila CRM sequences [31] enable a large-scale and more systematic study of the evolutionary patterns of CRM sequences . Such studies also crucially depend on accurate computational tools for sequence comparison . Commonly used multiple alignment tools [32]–[34] that treat regulatory sequences as no different from other types of DNA ( or for that matter amino acid ) sequences are known to be a source of errors in evolutionary analysis [35] , [36] . Even if the alignments are accurate , the step of annotating gaps as insertions or deletions ( usually done by ad hoc parsimony criteria ) may lead to inaccurate inferences [37] . We have previously developed new methods for inter-species sequence analysis , that are specially designed with the properties of regulatory sequences in mind . These include ( i ) Morph [38] , which optimizes pair-wise sequence alignment by using the known binding profiles of relevant transcription factors , and ( ii ) Indelign [39] , which uses a realistic probabilistic model of insertions and deletions to annotate “indel” events in a given multiple alignment . In this work , we take advantage of and extend these new methods to study the CRMs involved in Drosophila early development . This data set is ideally suited for such research because ( i ) the biological system is very well studied [8] and the relevant transcription factors are known , thereby limiting the false positives in binding site annotation , and ( ii ) much of the previous work on metazoan cis-regulatory evolution has been in this system [7] , [23] , [26] . Our study significantly extends the earlier work done on this dataset [40] and provides answers to many of the burning questions alluded to above .
We begin with our findings on the evolutionary behavior of transcription factor binding sites . We collected 68 D . melanogaster CRMs and seven TF motifs involved in the control of anterior-posterior segmentation in the blastoderm stage embryo . These CRMs ( source: REDfly [31] ) have been experimentally determined , without using evolutionary conservation for discovery , and are hence suitable for evolutionary studies without introducing ascertainment bias . Orthologous sequences of these CRMs were extracted from 11 other Drosophila species and were aligned by a special multiple alignment program , called “ProbconsMorph” . This is a new computational tool that we have developed , and is geared towards multiple alignments of regulatory modules in a TFBS-conscious manner ( see Methods ) . It avoids propagating pair-wise alignment errors to the entire multiple alignment by combining the “consistency transformation” ( see Methods ) of Probcons [41] with posterior alignment probabilities obtained from Morph [38] . We also repeated most of our tests using the alignment tool “Pecan” [42] that does not use TF motifs , and we point out differences , if any , between results from the two types of alignment . We annotated binding sites for each transcription factor , in the subset of D . melanogaster CRMs that overlap with ChIP-bound regions from Li et al . [43] , if such data was available . Site prediction was based on the p-value of match to the respective PWM ( “position weight matrix” ) motif . We contrasted the density of these binding site predictions ( in “bound” CRMs ) with those in “unbound” intronic sequences , and typically found 2–3 fold enrichment in the former . ( See Text S1 , “False positive proportion estimation” . ) We also predicted sites in each of the 11 other species separately , using the same method . Considering a binding site to be conserved if it is present in all other species in the D . melanogaster subgroup , we found that conserved sites were 2–3 fold enriched in CRMs than in intronic sequences . ( See Text S1 , “False positive proportion estimation” . ) Our findings are consistent with earlier results in Li et al . [43] , suggesting that the majority of predicted sites are likely to be functional . Binding sites from different species , that overlap each other in the multiple alignment , are collectively referred to as an “orthologous TFBS set” . Sites in such an orthologous set were re-aligned locally in order to correct for any errors in their precise alignment . Graphic visualizations ( Figure S1 ) of these 12-species CRM alignments , with binding site annotation , are available at our site http://europa . cs . uiuc . edu/TFBSevolution/ . Different positions in binding sites have different contributions to the binding affinity of the TF . Positions that form the core regions for TF-DNA binding are more specific ( less variation allowed ) in the motif , and should be under stronger selective constraints . We thus expect different positions of TFBSs to have different degrees of evolutionary conservation . The specificity of a position can be expressed by the information content ( IC ) of the corresponding column in the PWM ( position weight matrix ) , and the evolutionary rate by the number of substitutions in that position in orthologous binding sites ( see Methods ) . We observed highly significant negative correlations between specificity and evolutionary rate in five of seven TFs ( i . e . , all except Cad and Tll ) ( Table 1; Figure 1; Figure S2 ) . Thus , our results confirm earlier similar findings in bacteria , yeast and vertebrates [12] , [14] , [15] . To avoid a bias introduced by the use of PWM-guided alignments , we used Pecan alignments ( see Methods ) of five closely related species for this particular analysis . The results were reproduced when using ProbconsMorph alignments ( Table S1; Figure S3 ) . While substitution rates in a TFBS are position-specific , this does not imply that different positions evolve independently , although such an assumption is often made in existing evolutionary models [17] , [18] , [44] . It is easy to see that the exact same substitution can have drastically different effects on the functionality of a site , depending on how strong the site was to begin with . A site that is close to optimal will probably remain a site even if a crucial nucleotide is changed , thus this substitution is likely to be fixed . On the other hand , the same nucleotide change inside a weak site may have a larger functional consequence ( the site loses its binding functionality ) , thus will be less likely to be fixed . It therefore seems plausible that the substitution rate of a position should depend on the entire site . To study evolution at the level of binding sites , as opposed to nucleotides , we developed a simple mathematical model of binding site evolution , called “Site-level Selection” or “SS” model , that treats binding sites as single evolutionary units . Under this population genetics-based model , the fitness of a site can take two values , 1 if the binding affinity of this site is below some threshold , and if the affinity is above this threshold , for . ( We use the same threshold as that used for defining a binding site . ) The rate of substitution from site to , , is determined by the fitness difference between and according to this equation from population genetics theory [19] , [45]: ( 1 ) where is effective population size , is the mutation rate of to , and is the fitness function defined above . When , we have ; when , i . e . , there is a site gain , we apply the approximation that : ( 2 ) When , i . e . , there is a site loss , similarly we have: ( 3 ) Note that and are inseparable in the above equations , so we will use the single quantity 4Ns as measuring the intensity of selection . We tested how well this model fits the data on binding site evolution , and compared it to another model , called the “Halpern-Bruno” or “HB” model [17] , which assumes positional independence and purifying selection at each position of the TFBS . The HB model has been used previously in cis-regulatory analyses ( e . g . , Moses et al . [13] ) . We considered predicted binding sites in D . melanogaster and their respective aligned sequences ( whether designated binding site or not ) in a closely-related species ( D . yakuba ) , arbitrarily calling the former sites “ancestral” and the latter sites “descendant” . Assigning an “energy score” to each binding site based on its similarity to the PWM [46] , we calculated the difference in energy score between the ancestral and descendant sites , and used this as the statistic to represent binding site evolution . We computed , for each TF , the histogram of this “energy difference” statistic , and asked how well this histogram fits theoretical predictions from simulations using either the SS or the HB model ( Table 2 ) . For every motif , the SS model showed a significantly better fit to the data than the HB model . ( Table 2; Figure 2A; Figure S4 ) . ( See Methods for details of how statistical significance was estimated , while accounting for the additional free parameter in the SS model . The results were reproduced when using Pecan alignments; see Table S2 and Figure S5 . ) Our estimated level of selection ( 4Ns in the range 8–19 ) is consistent with an early estimate from bacterial regulatory sequences [19] and our results argue in favor of models treating entire binding sites as evolutionary units . However , in absolute terms , neither model explains the data very well ( Figure 2A; Figure S4 ) , and there is a greater amount of conservation ( energy differences close to zero ) in the observed data than predicted even with strong selection . A similar analysis was performed with the evolutionary statistic being the number of substitutions between ancestral and descendant sites , and we found that there is an excessive number of fully conserved sites ( no substitutions ) than expected under either the HB or the SS model ( Binomial test , p-value<10−12 ) ( Figure 2B; Figure S6; Figure S7 with Pecan alignments ) . This seems to indicate that for many sites , the allowed binding affinities fall in some narrower range , instead of being determined by a single threshold ( lower bound ) . It has been suggested that in order to produce the correct expression pattern , a binding site may prefer some specific affinity level , and both stronger and weaker binding tend to be less functionally optimal [47] . Our results provide support for this hypothesis . Even though TFBS loss and gain ( henceforth called “turnover” ) have been commonly observed , it is not clear whether these changes are adaptive [48] or not [26] . If adaptive selection is the main force behind binding site turnover , it is likely that the process will show a lineage-specific pattern; on the other hand , a molecular clock has been known to be suggestive of the absence of adaptive selection , as per the neutral theory of evolution [49] . We considered the fraction of binding sites in D . melanogaster that have an ortholog ( above threshold ) in a second species , and plotted this fraction as a function of evolutionary divergence from the second species ( Figure 3; Figure S8 ) . For all transcription factors , the fraction of shared binding sites decreases linearly ( R2>0 . 90 , Table 3 ) as the divergence time increases , a clear sign of a molecular clock . One problem that may confound the analysis is the presence of false positive binding sites predictions , which are expected to follow a molecular clock . To examine this effect , we calculated a correction term in the fraction of conserved sites , and regressed this with divergence time , using the false positive proportion as a free parameter . High values of the adjusted R2 were obtained ( Table 3 ) , confirming the presence of the molecular clock . We repeated the exercise with sites for randomly created PWMs , and found a similar linear relationship . The rate of loss ( negative slope of the line ) for these random sites is higher than the rates for binding sites , for six of the seven transcription factors ( Table 4 ) , the difference being significant for Bcd and Kr . We note that the sites predicted by random PWMs do not represent neutral sequences , but reflect the average constraint in CRM sequences . This has been shown previously in [43] . The results were reproduced when using Pecan alignments ( Table S3 and S4; Figure S9 ) . Having characterized some general patterns of TFBS evolution , in this section we study what specific factors may influence the conservation and turnover of binding sites . Finally , we analyzed insertions and deletions in known regulatory sequences , to study the extent of indel-purifying selection . Among 370 non-overlapping D . melanogaster CRMs from the REDfly database [31] , we chose 128 CRMs that have clear orthologous sequences in D . simulans , D . yakuba , and D . erecta . This choice of species was dictated by simulation-based assessment of the limits of our indel annotation capability ( see Methods ) . Because insertions and deletions ( indels ) may have different functional consequences on CRMs , we treat them differently . We estimated the number of short insertions and deletions in CRMs using Pecan [42] for alignment and Indelign [39] to annotate the indels . For each CRM , the insertion or deletion count was defined as the average of the respective counts in the four species , weighted by the branch length . We compared indel frequencies in CRMs to those in “background sequences” , chosen to be the regions flanking the CRMs . We found that ( i ) the number of short deletions ( less than 20 bp in length ) in CRMs is significantly smaller than that in background regions ( paired Wilcoxon signed-rank test , p-value 0 . 0074; 1970 in CRMs and 2183 in length-matched background regions ) and ( ii ) there was no statistically significant difference ( p-value 0 . 5464 ) in the number of insertions ( 1932 in CRMs and 1870 in background ) . The number of long indel events ( 20 bp or longer ) in our data set was relatively small ( CRM: 107 insertions and 175 deletions , background: 115 insertions and 178 deletions ) and no significant difference was observed in this regard between CRMs and background regions . Another related question is the indel pattern in the “spacer” region between CRMs and transcription start site ( TSS ) of the target genes . Transcriptional regulation depends on the communication between CRMs and promoter sequences [54] , which may pose some requirements on the length of the spacer sequences . We thus repeated the above analysis on these spacer regions . ( We only consider 63 upstream CRMs in this experiment . ) No significant differences in frequencies of insertions or deletions were observed between these regions and background sequences ( data not shown ) . Our results show that indel-purifying selection exists on CRM sequences , but such selection acts most strongly on deletions . We did not find clear suppression of long-indels , as has been observed before [27] .
The study of cis-regulatory evolutionary patterns has provided important insights on regulatory sequence function [23] , [55] , and proves valuable for prediction of these sequences in genomes [9] , [56] . Yet , our understanding of cis-regulatory evolution is limited at best . While we have theories as well as a large volume of empirical data on protein evolution , we essentially have no theory and have made limited observations on the evolution of regulatory sequences . Our goal here is to begin to bridge the gulf between the vast amount of genomic sequence data and our poor understanding of regulatory sequences and their evolution . We have conducted a detailed evolutionary analysis of a large collection of experimentally verified CRM sequences , taking advantage of the recently sequenced 12 Drosophila genomes . Our analysis has revealed several interesting patterns , some along expected lines ( but not confirmed previously ) , and some contrary to our expectations . We believe that our work will furnish evidence orthogonal to experimental characterization for understanding the organizational principles of CRMs , and will be important for developing a theory of regulatory evolution in the future . There are several technical issues that were important to address in our analysis . Evolutionary comparison depends on the alignment of orthologous sequences , but in general , alignments cannot be perfectly determined and may be a source of biased conclusion [36] . This may be a particularly serious problem for the analysis using 12 Drosophila species because of the relatively large divergence . We addressed this concern by developing a new multiple alignment program tailor-made for regulatory sequences . It combines the power of a pair-wise regulatory sequence alignment tool , Morph [38] , and a probabilistic multiple alignment framework Probcons [41] . We have made this new software ( ProbconsMorph ) available freely for public use , to facilitate future studies of this genre . Nevertheless , the use of motifs to construct alignment may artificially boost the conservation level of TFBSs . We carefully addressed this potential bias whenever it may affect our conclusion . For example , when testing the positional variation of substitution rates , we use Pecan-based alignments without using motifs and limited ourselves to five closely related species . Similarly , when testing the correlation of binding site strength to turnover rates , we use randomized PWMs ( as “negative controls” ) to validate our finding . We also repeated all our analyses with Pecan-based alignments . The various trends seen in Results were almost always reproduced . One notable difference was that the correlation between nearest homotypic site distance and evolutionary rate ( Table 6 ) for Cad was statistically significant ( p-value 0 . 02 ) in ProbconsMorph alignments , but insignificant ( p-value 0 . 15 ) in Pecan alignments . We suspect that this may be due to the tendency of standard alignment tools ( such as Pecan ) to misalign one or two nucleotides at the boundary of binding sites , especially if the motif contains short repeats such as TTTT [38] , as is the case for Cad . Another critical component of our analysis is the prediction of TFBSs . By using the same PWMs for all the genomes , we have made the assumption that the PWM of any TF is fully conserved across 12 Drosophila genomes . This is questionable , as researchers have found in yeast that the change of TF binding specificities can be an important part of the evolutionary change of regulatory networks [57] . For the seven motifs we analyzed , however , there is prior computational evidence that the binding specificities have not changed between D . melanogaster and D . pseudoobscura [58] . Another issue related to TFBS prediction is that predicted binding sites tend to have a high proportion of false positives [59] . We believe this problem is mitigated by our focus on the segmentation network , the fact that we restrict ourselves to transcription factors and CRMs experimentally known to be involved in regulating the segmentation genes , and our use of ChIP-based binding information wherever possible . We also believe that within a CRM , any computationally predicted binding site for a relevant transcription factor can “attract” transcription factor molecules , and contribute to the expression pattern , and should thus be considered “functional” in a broad sense . The results from Janssens et al . [60] seem to support this point . In practice , we may still have a small number of false predictions because of inaccuracies of the PWMs and we have attempted to estimate the false positive proportion by various methods ( see Text S1 ) . Also note that while false site predictions may obscure the evolutionary pattern of functional binding sites , they will not , in general , introduce spurious patterns ( since , by definition , these sites are not under selection ) . In cases where the false sites may affect our interpretation of results , for example , in the test of molecular clock for binding site turnover , we have tried to make appropriate corrections . In addition , in estimating TFBS turnover rates , we have emphasized on losses rather than gains of sites , because a predicted TFBS loss event has stronger supporting evidence than a gain event ( the “gained” site is more likely to be a false positive prediction ) . Our model of binding site evolution , the “Site-level Selection” ( SS ) model , is a special case of the population genetic model proposed by Mustonen and Lassig [19] . Under their model , the fitness of a site is determined by its binding energy . The difference of the energy distribution of known sites and of the neutral sites allows one to estimate the fitness of any energy value . A binding site evolves in the space of all possible sequences , with the transition rate between any two sequences determined by the fitness values of the two sequences , given by Equation ( 1 ) . For most known TFs , however , the number of known sites is too small to reliably estimate a fitness function and the simplification introduced in our model is probably necessary . Our SS model is also similar to the model in Raijman et al [61] . Under this model , a site always tends to preserve its current functional status , that is , the substitution in a binding site that makes is nonfunctional will have a lower rate , and similarly , a substitution that creates a functional site in an originally neutral site will also have a lower rate . However , their model is not formulated in population genetic terms and the transition from a non-site to site is always selected against ( this will be favored under the Mustonen-Lassig model and ours ) . We found that the SS model better explains the evolutionary pattern of binding sites than the HB model , which assumes the independence of substitutions at different positions of a site . A recent study [62] also reported this dependence of binding site positions , though without directly comparing two kinds of models . Admittedly , the presence of false sites may complicate our analysis . It is difficult to directly address this issue , say , through a mixture model approach as done in [22] because of the difficulty of computing probabilities under the SS model . However , we note that if we were to remove false sites from the observed data , we would see a greater proportion of conserved sites , implying that the SS model will continue to be closer to the observation than the HB model ( see Figure 2 ) . Next , we observe an overrepresentation of fully conserved sites ( no mutations ) compared to what is expected from both SS and HB models ( Figure 2B ) . This argues for the conservation of precise affinities , a hypothesis consistent with our current knowledge about the dependence of expression pattern on precise binding affinities [63] , [64] , though this phenomenon has not been statistically observed previously . Finally , we note that the findings of position-specific substitution rates and site-level selection are not contradictory; as pointed out in [19] , each position of the site contributes separately to the fitness of the site , which depends on the sum-total of these contributions . Our findings of a molecular clock extend earlier results on a small number of well characterized CRMs [65] across three Drosophila species , suggesting that this is a property common to developmental CRMs across a large evolutionary range . Even though we cannot exclude the presence of adaptive selection in individual cases , our results seem to suggest that negative selection to maintain the existing binding sites is the dominant mode of evolution , coupled with the occasional loss of sites due to random drift . The rate of site loss likely reflects the strength of purifying selection . Our tests point out that stronger binding sites are conserved more often than weaker sites . This is consistent with an earlier study [66] , which found that stronger Dorsal binding sites were more likely to reside in conserved blocks . A simple explanation for this is that stronger sites are more likely to be important to CRM function , thus under stronger constraint . An alternative explanation is that there is a “quality” threshold that defines functionality and once a site drops below that threshold , it is impervious to selective forces . Assuming this is true , we note that a weaker site is closer to the threshold than a stronger site , and may thus be lost more easily . A recent paper [61] seems to support the latter hypothesis . It is likely that the forces of natural selection as well as those of mutation/random drift together determine the evolutionary fate of a binding site , as suggested by Mustonen and Lassig [19] . An unexpected result of our analyses is that the degree of homotypic clustering does not affect turnover rate . This is contrary to the notion that more binding sites of the same type will lead to greater redundancy , easing the selective pressure on the individual sites . Instead , the number of binding sites seems to be important to CRM function . This observation is similar to one of the implications of our findings of site-level selection: that exact affinities of binding sites are functionally important . Both observations are consistent with the so called “gradient threshold model” [47] , which suggests that different genes may respond to different concentration levels of the same TF by harnessing different numbers of binding sites with varying affinities . The exact binding affinities and number of sites are important under this model . In a more detailed analysis of homotypic clustering , now considering the binding site arrangement , we observed that for some factors , if a site is adjacent to another site of the same factor , this site will be less likely to be lost during evolution . This may be indicative of cooperative activity of proximal homotypic binding sites , leading to stronger selective pressure . For instance , the significant result ( p-value 0 . 0184 , Table 6 ) for Cad is consistent with anecdotal evidence of Cad sites being located as proximal pairs [67]–[69] , although we are not aware of any biochemical evidence for such cooperativity . There is also some evidence in the literature for DNA binding by homodimers of Tll [70] and Hb [71] . Our observation also suggests that sites that have a proximal “partner” are perhaps less likely to be spurious sites , which will provide a useful additional guideline to binding site prediction [72] . Surprisingly , we did not observe significant result for Bcd , even though it is known to bind cooperatively [52] . This negative result is a reminder that the sensitivity of our statistical tests may be reduced due to a variety of factors , e . g . , alignment errors , false sites , etc . These factors are unlikely , however , to produce spurious statistical signals . We found that the presence of a binding site for a different factor , either overlapping or proximal to a binding site , can strongly affect the latter's evolution . Different mechanisms of local interactions between sites are known in developmental CRMs , e . g . , cooperative binding between two factors [73] , [74] , short-range quenching [75] , [76] , competitive binding to overlapping sites [74] , etc . In all these cases , the loss of a single binding site may disrupt the interaction and create a larger change of expression than if the binding sites act in an additive fashion . As a consequence , these locally interacting site pairs may be under stronger selection . Our results support the importance of context in determining evolutionary fate of binding sites . A recent paper reports similar results for four CRMs of the even-skipped gene [53] . By working on a much larger set of CRMs , we confirm this context-dependence as a general evolutionary pattern . We also found some interesting specific cases , for example , the Kr sites that overlap with another TF site , appear more conserved , consistent with the known role of Kr as a repressor with the ability of competitive binding . In addition , the difference of the evolutionary patterns of the seven TFs suggests that they may depend on different mechanisms for their function . For example , both Kr and Tll are repressors , but Tll is more conserved if it is adjacent to some other site , while Kr is more conserved if it overlaps with another site . This seems to suggest that the relative importance of competitive binding and short-range quenching may be different in Kr and Tll . We did not find strong evidence of suppression of large indels within CRMs relative to their flanking sequences . Our results are different from an earlier study of indel patterns of CRMs in sea urchins , which reports that large indels ( >20 bp in length ) are virtually absent inside CRM sequences [27] . There is an alternative explanation for this discrepancy: it has been known that Drosophila has a very compact genome as the neutral deletion rate is very high [77] and a large fraction ( 40–50% from different estimates ) of intergenic non-coding sequences is under evolutionary constraint [48] , [78] . Consequently , the flanking sequences of CRMs may not be entirely neutral , and the distinction between CRM and flanking sequences may not be as pronounced as in other species . ( Our options were limited with respect to the “background” sequence to contrast with , since long repeats often used as neutral sequence in mammalian genomes [79] are rare in Drosophila . ) The fact that short deletions are more constrained than short insertions is likely due to different effects of insertions and deletions on CRM sequences: any deletions that extend to an existing binding site will annul its functionality , while insertions , unless occurring exactly inside TFBSs , will only change the distance between sites , but not destroy them . In Text S1 , we outline an illustrative calculation , suggesting that under simple but reasonable assumptions , short deletions are maybe twice as more likely to interfere with a binding site than are short insertions . These results combined with the lack of strong constraint on spacer sequences suggest that CRM structure is overall flexible , permits relatively quick evolutionary change , and functions without being very sensitive to the precise distances between binding sites . In terms of its implications for bioinformatics , our results seem to indicate that the indel signature can be a useful CRM predictor but not strong enough to work alone , somewhat contrary to prior expectations [27] , [28] .
12 Drosophila genome sequences from D . ananassae ( Feb . 2006 assembly ) , D . erecta ( Feb . 2006 assembly ) , D . grimshawi ( Feb . 2006 assembly ) , D . melanogaster ( Apr . 2006 assembly , release 5 ) , D . mojavensis ( Feb . 2006 assembly ) , D . persimilis ( Oct . 2005 assembly ) , D . pseudoobscura ( Feb . 2006 assembly ) , D . sechellia ( Oct . 2005 assembly ) , D . simulans ( Apr . 2005 assembly ) , D . virilis ( Feb . 2006 assembly ) , D . willistoni ( Feb . 2006 assembly ) , and D . simulans ( Nov . 2005 assembly ) were compiled from UCSC Genome Browser database [80] . To predict the positions of putative TFBSs , position weight matrices ( PWMs ) for seven TFs , Bcd ( Bicoid ) , Cad ( Caudal ) , Dstat , Hb ( Hunchback ) , Kni ( Knirps ) , Kr ( Kruppel ) , and Tll ( Tailless ) were compiled from FlyReg [81] and the literature . We used the phylogenetic tree and branch lengths for the 12 species in [82] and for the four species ( D . melanogaster , D . simulans , D . yakuba , and D . erecta ) in [25] . Orthologous sequences of each D . melanogaster CRM were obtained by the liftOver program from the UCSC Genome Browser database . The background region corresponding to a CRM was defined as the region upstream of the farthest known CRM of its target gene , equal in length to its corresponding CRM . For the analysis of TFBS evolution , we developed a new multiple alignment program , “ProbconsMorph” , by integrating Probcons [41] , a consistency based multiple sequence alignment program , and Morph [38] , a pair-wise sequence alignment program that is specially designed to align regulatory modules . Morph uses a pair-HMM as a generative model for alignment of two orthologous CRMs , and is parameterized by the given motifs , as well as various evolutionary rate parameters that it fits to the data . It uses maximum likelihood inference to simultaneously perform TFBS annotation and alignment . It reports for every pair of positions in the two sequences , the posterior probability that they are aligned . Morph was run to produce such a probabilistic alignment of every pair of species . Probcons takes such pair-wise alignment probabilities and builds a multiple sequence alignment progressively , while using the “consistency transformation”: the probability of alignment of two nucleotides and is updated based on the alignment probabilities of and and of and , where is a nucleotide from a third species . We have shown previously that Morph provides practical benefits for inference of evolutionary events and rates by computing a better alignment; ProbconsMorph is an effective and efficient extension of this program to more than two species . We made two simple modifications to Probcons to integrate it with Morph: firstly , Probcons was made to work on DNA sequences ( the current implementation handles protein sequences only ) , and secondly , it was made to accept a phylogenetic tree as input , rather than estimate the tree at run-time . The ProbconsMorph software is publicly available at our site http://europa . cs . uiuc . edu/TFBSevolution/ . Pecan [42] was used for the alignment of four species in the analysis of indels in CRMs and spacers . We have performed extensive studies on simulated data to determine the limits of indel annotation , and estimated that accurate labeling of insertions and deletions is only possible for the four closely related species D . melanogaster , D . simulans , D . yakuba , and D . erecta . ( Kim and Sinha , in preparation . ) Pecan alignments of these four species , and D . sechellia , were also used for the study of position-specific substitution rates in binding sites ( Table 1 ) . Insertion and deletion annotations were done using our previously published Indelign program [39] that is based on a probabilistic model of indels and annotates indels as being insertions or deletions based on maximum likelihood . We used a mixture of two geometric distributions as a model of the length distribution of indels . As shown in Figure S10 , this mixture model is a much better fit to the indel length distributions empirically observed in D . melanogaster CRMs used in this study and their orthologous sequences in D . simulans , D . yakuba , and D . erecta . The new version of the Indelign program is available at our site http://europa . cs . uiuc . edu/TFBSevolution/ .
|
The spatial–temporal expression pattern of a gene , which is crucial to its function , is controlled by cis-regulatory DNA sequences . Forming the basic units of regulatory sequences are transcription factor binding sites , often organized into larger modules that determine gene expression in response to combinatorial environmental signals . Understanding the conservation and change of regulatory sequences is critical to our knowledge of the unity as well as diversity of animal development and phenotypes . In this paper , we study the evolution of sequences involved in the regulation of body patterning in the Drosophila embryo . We find that mutations of nucleotides within a binding site are constrained by evolutionary forces to preserve the site's binding affinity to the cognate transcription factor . Functional binding sites are frequently destroyed during evolution and the rate of loss across evolutionary spans is roughly constant . We also find that the evolutionary fate of a site strongly depends on its context; a pair of interacting sites are more likely to survive mutational forces than isolated sites . Together , these findings provide new insights and pose new challenges to our understanding of cis-regulatory sequences and their evolution .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genetics",
"and",
"genomics/comparative",
"genomics",
"computational",
"biology/transcriptional",
"regulation",
"computational",
"biology/evolutionary",
"modeling",
"evolutionary",
"biology/pattern",
"formation",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2009
|
Evolution of Regulatory Sequences in 12 Drosophila Species
|
Melioidosis is an infectious disease that is transmitted mainly through contact with contaminated soil or water , and exhibits marked seasonality in most settings , including Southeast Asia . In this study , we used mathematical modelling to examine the impacts of such demographic changes on melioidosis incidence , and to predict the disease burden in a developing country such as Thailand . A melioidosis infection model was constructed which included demographic data , diabetes mellitus ( DM ) prevalence , and melioidosis disease processes . The model was fitted to reported melioidosis incidence in Thailand by age , sex , and geographical area , between 2008 and 2015 , using a Bayesian Markov Chain Monte Carlo ( MCMC ) approach . The model was then used to predict the disease burden and future trends of melioidosis incidence in Thailand . Our model predicted two-fold higher incidence rates of melioidosis compared with national surveillance data from 2015 . The estimated incidence rates among males were two-fold greater than those in females . Furthermore , the melioidosis incidence rates in the Northeast region population , and among the transient population , were more than double compared to the non-Northeast region population . The highest incidence rates occurred in males aged 45–59 years old for all regions . The average incidence rate of melioidosis between 2005 and 2035 was predicted to be 11 . 42 to 12 . 78 per 100 , 000 population per year , with a slightly increasing trend . Overall , it was estimated that about half of all cases of melioidosis were symptomatic . In addition , the model suggested a greater susceptibility to melioidosis in diabetic compared with non-diabetic individuals . The increasing trend of melioidosis incidence rates was significantly higher among working-age Northeast and transient populations , males aged ≥45 years old , and diabetic individuals . Targeted intervention strategies , such as health education and awareness raising initiatives , should be implemented on high-risk groups , such as those living in the Northeast region , and the seasonally transient population .
Melioidosis is an infection caused by the Gram-negative bacillus Burkholderia pseudomallei , which exhibits marked seasonality in most settings where it is endemic , including Southeast Asia and Northern Australia [1] . Melioidosis is a communicable disease that is usually transmitted via contaminated soil or water , and is highly prevalent in Northeast Thailand [2] . Most of the population at risk of melioidosis lives in rural areas , especially those people who frequently come into contact with soil or water , such as rice farmers [3 , 4] . In Thailand , the highest number of melioidosis reported cases are often in January and October [5] . Infection with B . pseudomallei shows great clinical diversity , spanning asymptomatic infections , localized skin ulcers or abscesses , chronic pneumonia mimicking tuberculosis , and fulminant septic shock with abscesses in multiple internal organs [6] . Both humans and animals are susceptible to B . pseudomallei , and may be infected by percutaneous inoculation , inhalation , or ingestion . Person-to-person spread and zoonotic infections of humans are very rare [7] . The incubation period is between 1–21 days ( average 9 days ) [8] , and is believed to be influenced by the inoculation dose , mode of infection , host risk factors , and probably differential virulence of the infecting organisms . Most cases result from recent infections , although latency with reactivation has been described up to 62 years following exposure [8] , while the median times to relapse and reinfection are 21 weeks and 111 weeks , respectively . The risk of relapse is related to a patient’s adherence to treatment and the initial extent of disease , but not to any underlying conditions [9–11] . Melioidosis seems to be more severe in older people with lower immunity or chronic underlying conditions , such as diabetes [12] . The risk of contracting melioidosis in diabetic individuals is 12 times higher than for non-diabetic individuals [13 , 14] . Currently , the global burden of melioidosis is estimated to be 165 , 000 cases per year ( 95% credible interval 68 , 000–412 , 000 ) , with 89 , 000 deaths ( 36 , 000–227 , 000 ) [15] . Thailand’s Bureau of Epidemiology ( BoE ) launched a melioidosis surveillance system in 2001 ( Report 506 ) [5] . Approximately 80% of reported melioidosis cases were from Northeast Thailand [5] . In the past , the number of cases shown in the surveillance system was heavily relied on provincial and regional hospitals voluntarily report , very few were reported from private hospitals [16] . In general , melioidosis is diagnosed by testing for antibodies to B . pseudomallei using an indirect hemagglutination ( IHA ) technique , which has been found to have low sensitivity and specificity [17] . This surveillance system was revised in 2010 in order to capture more health data items . There has been an increase in usage of bacterial culture [16] which could give rise to an increase in total number of culture-confirmed cases . In addition , there has been an improvement to access to healthcare . Nevertheless , the true number of cases is still under-reported because of diverse clinical manifestations and inadequate bacterial identification methods . A previous estimation suggested cases in Thailand were in excess of 7 , 000 cases per year [15] , while the BoE reported just 3 , 242 cases in 2015 [5] . B . pseudomallei is resistant to a wide range of antimicrobials , and ineffective treatment may result in death in 70% of cases [18] . The treatment for melioidosis consists of an intensive phase of at least 10–14 days of ceftazidime , meropenem , or imipenem , administered intravenously , followed by oral eradication therapy , usually with trimethoprim–sulfamethoxazole ( TMP-SMX ) for 3–6 months [19] . There is currently no vaccine against melioidosis [20 , 21] . The demographics of Thailand are currently in a transition phase , becoming more like those of developed countries , with rapid changes in population structure , reductions in birth and mortality rates , and a low rate of population growth . Urbanization is accelerating , and there are large annual population movements . These types of changes have been shown to have important impacts on public health and the disease burden of both non-communicable [22] and communicable diseases [23] . The population at highest risk of contracting melioidosis is the working age group . There is appreciable seasonal movement among this group as they go about their working lives . The internal migration of Thai people involves a number of distinct forms of movement within each year . Three forms have been identified in previous research [24]: a single movement , seasonal movement , and repeated movement . Seasonal migration involves people moving from the North and Northeast regions of Thailand towards the Bangkok metropolis and the Central region during the dry season ( from March through to June ) , and in the reverse direction during the wet season ( June to September ) [24] . 40% of people from the Northeast are classified as seasonal migrants ( a transient population ) [25] . It is obvious that for person-to-person transmissible infections , there are significantly more infections when such transient individuals are considered [26–29] . However , very few studies were trying to look at the effect of transient populations on an infectious disease from a primarily environmental source which will help better describe the temporal and spatial changes of the incidence of such a disease [30] . Developed countries are also observing an emergence of melioidosis related to travelling and importation of cases [1] . To date , only a few approaches have been applied to determine the melioidosis burden , including simple maps of melioidosis [1] , maps of the global distribution of B . pseudomallei , and estimates of the total incidence and mortality due to melioidosis worldwide using a statistical model [15] . Only one study has used mathematical modelling , exploring the use of childhood seroprevalence data as a marker of intensity of exposure [31] . In this study , we used mathematical modelling to predict the incidence of melioidosis in the Thai population , taking account of population changes , seasonal movement , and incidence of diabetes . The model provides multi-dimensional forecasting of melioidosis , which could be useful for targeting intervention strategies in this setting .
We generated a deterministic demographic sub-model to predict the size of the total population ( see S1 Figure A ) . We stratified the population by age and sex into 100 annual interval classes , from 0 to 100 years old . The population in each class followed the actual population structure of Thailand between 1980 and 2015 , based on birth , death , and migration rate data from the Population and Housing Census [32 , 33] , and using the 1980 census data as the initial condition . All females in the age classes between 15 to 50 years old were considered to be capable of reproduction , with the fertility rate ( fr ) [34] , while the death rate was age-related [35] . Members of the population were assumed to die upon reaching 100 years of age . Crude net migration rates ( immigrant minus emigrant per 1 , 000 population ) for each year had an impact on all age and sex compartments [36] . Most of the at-risk population for melioidosis lives in rural areas , especially in Northeast Thailand , so we modelled internal migration by classifying the population of Thailand into three independent groups . These were: those from the Northeast region who live at home for more than 6 months in a year ( NE ) , the transient group or the people from the Northeast region who move seasonally between home and other parts of the country and spend less than 6 months in a year at home ( T ) and lastly the non-Northeast group , who live somewhere other than the Northeast ( Non_NE ) . We created the seasonal movement sub-model to overlay with the demographic sub-model to estimate the rates of movement among them ( see S1 Figure B ) . We solved a large set of ordinary differential equations ( ODE ) for the deterministic demographic sub-model and the seasonal movement sub-model , defined in S1 Information on Demographic sub-model and Seasonal movement sub-model , respectively . The demographic and seasonal movement sub-model was overlaid with the melioidosis infection model , defined in S1 Information on Melioidosis infection sub-model . In the melioidosis infection model ( a susceptible , exposed , infected , recovered , susceptible , or SEIRS , model ) , the population was further divided into eight health compartments: susceptible ( S ) , diabetic susceptible ( SDM ) , exposed ( E ) , symptomatic ( Sym ) , asymptomatic ( Asym ) , severe ( Sev ) , treatment ( Treat ) , and recovered ( R ) ( see Fig 1 ) . Melioidosis case data stratified by age , sex , and geographical area were obtained from the Thai annual epidemiological surveillance reports from 2008 to 2015 [5] . Key assumptions for our model were as follows . First , the transient population data used within this model referred only to the movement of the Thai population . The movement of migrant workers from other countries could be significant but was omitted in this study for simplicity [24] . Second , diabetes progression was assumed to be irreversible , i . e . people could not move from diabetic to non-diabetic . Third , we did not consider pre-diabetes or impaired glucose tolerance . Fourth , we assumed that incidence rates of diabetes were constant over time but varied by age . Fifth , we did not focus on chronic symptoms ( those of duration greater than two months ) , including such presentations as chronic skin infections , chronic lung nodules , or pneumonia , which only accounted for around 10% of melioidosis patients [12] . Finally , we did not focus on any behavioral factors such as excessive alcohol use . We used R software version 3 . 3 . 3 to run and analyze the model outputs , and the deSolve package to solve the differential equations [37] . The initial parameter values were calculated from population data and disease burden . Model fitting was carried out using the Markov Chain Monte Carlo ( MCMC ) method , implemented with the Bayesian Tools R package as defined in S2 Information on the Bayesian framework [38] . The demographic and seasonal movement sub-models were run from 1980 ( see S2 Figure A ) to calibrate the model by fitting to the average migration data , including the population in the Northeast moving to non-Northeast , and the reverse direction from 2005 to 2015 [25] . We estimated seasonal movement parameters from the transient population model ( see S1 Table A ) and used them to run the melioidosis infection model from 2005 . The model was run and fitted to the annual incidence of melioidosis by age , sex , and area by year , and seasonally by month , from 2008 to 2015 [5] . For model fitting , the DEzs method in the Bayesian Tools package allowed automatic parallelization on three cores to be used for sampling . This method allowed fewer chains to be used for estimated a large number of parameters and thus optimized the computational time [39] . Number of iterations and burn in were decided upon the model convergence by analyzing the differences between multiple Markov chains . The convergence was assessed by several measures including the standard procedure of Gelmal-Rubin [40 , 41] and the target acceptance rates [42] . Thirty-three parameters were estimated and the median values and credibility intervals were reported . These parameters were those representing the infection rates among both sexes in the Northeast , transient , and non-Northeast populations , ( βaNE , βaT , βaN ) respectively , proportion of symptomatic cases ( pE ) , recovery rate from asymptomatic ( σ ) , recovery rate from symptomatic ( γ ) , Relative susceptibility to melioidosis among diabetic individuals when compare with non-diabetic ( q ) , mortality/death rate for melioidosis ( μM ) , amplitude ( Ainc ) , phase angle ( φinc ) and proportion of reporting ( Report ) ( see S1 Table A ) . Note that the proportion ( 1- Report ) was defined as “Under-reporting” i . e . those symptomatic melioidosis patients that have been seen by a physician , but the physician did not report them to the public health authority for some reasons e . g . improperly diagnosis or missing report . The model was further used to predict the 20-year age-specific incidence of melioidosis among males and females in Thailand , sampling all 33 parameters from the posterior chains . The model predictions were reported as age , gender , and area-specific incidence rates over time .
The demographic sub-model was able to reproduce the past population structure of Thailand from 1980 to the present ( see S2 Figure A ) . The parameters that characterized seasonal movement were estimated by fitting the model to the population movement data ( see S2 Figure B ) . The model showed that majority of movements were made by Northeast individuals who moved to non-Northeast areas , approximately 13 , 600 persons per 100 , 000 population per month , or 34% of all movements within a month ( see S1 Table A ) . Moreover , the majority of movements were among those aged between 15 and 60 years old , about 19 , 000 persons per 100 , 000 population per month , or 51% of all movements within a month ( see S2 Figure C ) . The fitting performance is shown in Fig 2 . Melioidosis cases occurred seasonally , with a peak in the wet season that lasted from May to October . The infection parameters that minimized the fit statistic , using the Bayesian method , are shown in Table 1 . The highest infection rate was estimated to be 6 cases per 100 , 000 population per month among males aged 45–59 years old in the Northeast . The lowest rate was 0 . 4 cases per 100 , 000 population per month among females aged 15–44 years old in the non-Northeast region . Surprisingly , we found that the infection rate among the transient male population aged 15–44 years was higher than the non-Northeast population ( 0 . 8 compared with 0 . 08 per 100 , 000 persons per month ) . Overall 46% of melioidosis cases were symptomatic . Recovery rates for untreated symptomatic cases and asymptomatic patients were estimated by the model , with the average period of infection estimated at around 9 and 5 months , respectively . The susceptibility to melioidosis among DM population is 10 . 84 [95% CI 8 . 42–12 . 23] times greater than the non-DM population . If patients’ treatment failed and they developed severe melioidosis , they could die within two weeks . We estimated 80% and 50% under-reporting of cases in 2008–2009 and 2010–2015 , respectively . Projections of the numbers of melioidosis cases between 2015 and 2035 are given in Fig 3 . Total melioidosis incidence per year was projected to increase by 70% , from 6 , 569 ( 4 , 834–8 , 701 ) in 2015 to 11 , 173 ( 8 , 207–14 , 773 ) in 2035 . The largest increase of melioidosis was projected to occur among the population aged 45–59 years old . The predicted incidence among males was two-fold greater than that of females . The majority of melioidosis cases were seen to occur in the population from the Northeast region of Thailand . The predicted incidence among non-diabetic was two-fold greater than that of diabetic population . In Fig 4 , total melioidosis incidence rates were projected to increase by approximately 10% by 2035 , from 11 . 42 ( 8 . 5–13 . 4 ) in 2015 to 12 . 78 ( 9 . 6–14 . 9 ) per 100 , 000 population in 2035 ( see Table 2 ) . The highest incidence rates were predicted to be among those aged between 45–59 years old , followed by those age 60 years old and above . The incidence was almost double among males compared with females in both Northeast and other regions . The incidence rate among the Northeast population was more than double compared with the transient population , and almost ten times higher when compared with the other regions . This study also highlighted the importance of melioidosis among the transient population who temporally live in the risk area but had almost six times higher incidence compared with other regional populations . From diabetes prospective , the incidence of melioidosis among diabetes was predicted to be as high as 60 per 100 , 000 population . To summary , the risk of melioidosis among the aging population with some chronic diseases such as diabetes is presenting an increasing trend . The risk of infection among transient population , who temporary get some disease exposure during the agricultural seasons , cannot be ignored .
Population dynamics , seasonal movement , melioidosis infection rates , and under-reporting are important components of melioidosis incidence patterns . The increases seen in melioidosis cases are partly attributable to demographic changes as working , transient , and aging population groups are more prone to develop melioidosis . The key findings from our study are firstly , the increasing trend of melioidosis incidence , especially among males aged 45–59 years old , is predicted to continue; and secondly , the male , Northeast , and transient populations aged 45–59 years old were at the highest risk of melioidosis infection . We anticipate that the modelling methods described here could be used in similar settings , especially those with reliable census data , to estimate the future melioidosis burden , as well as the potential effects of under-reporting . In addition , this modelling approach could be adapted to study other infectious diseases , behavioral changes , and seasonal movements , where demographic factors are important drivers of a population’s disease burden .
|
Melioidosis is an infectious disease caused by the Gram-negative bacillus Burkholderia pseudomallei , which exhibits marked seasonality in most settings where it occurs , such as Southeast Asia and Northern Australia . Most of the population at risk of contracting melioidosis lives in rural areas; particularly at risk are those who are exposed to soil and water , such as rice farmers . Thailand’s demography is in a transient phase , with older age groups set to double within a decade . Social impacts of lifestyle changes are reflected in seasonal movement and increasing urbanization . In this study , we used mathematical modelling to examine the impacts of such demographical changes on an important infectious disease and to dynamically predict the disease burden in a developing country setting , namely Thailand . We found that melioidosis incidence was significantly higher among working-age Northeast and transient populations , specifically among males aged ≥45 years old and individuals with diabetes . Improved health education and awareness raising should be implemented on a national scale , with a focus on high-risk groups living in endemic areas , as well as those who move seasonally between these and other areas .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"death",
"rates",
"medicine",
"and",
"health",
"sciences",
"melioidosis",
"population",
"dynamics",
"geographical",
"locations",
"diabetes",
"mellitus",
"bacterial",
"diseases",
"age",
"groups",
"endocrine",
"disorders",
"population",
"biology",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"endocrinology",
"metabolic",
"disorders",
"people",
"and",
"places",
"population",
"metrics",
"asia",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"thailand"
] |
2019
|
Modelling population dynamics and seasonal movement to assess and predict the burden of melioidosis
|
The human sex chromosomes differ in sequence , except for the pseudoautosomal regions ( PAR ) at the terminus of the short and the long arms , denoted as PAR1 and PAR2 . The boundary between PAR1 and the unique X and Y sequences was established during the divergence of the great apes . During a copy number variation screen , we noted a paternally inherited chromosome X duplication in 15 independent families . Subsequent genomic analysis demonstrated that an insertional translocation of X chromosomal sequence into theMa Y chromosome generates an extended PAR . The insertion is generated by non-allelic homologous recombination between a 548 bp LTR6B repeat within the Y chromosome PAR1 and a second LTR6B repeat located 105 kb from the PAR boundary on the X chromosome . The identification of the reciprocal deletion on the X chromosome in one family and the occurrence of the variant in different chromosome Y haplogroups demonstrate this is a recurrent genomic rearrangement in the human population . This finding represents a novel mechanism shaping sex chromosomal evolution .
The human sex chromosomes originate from an ancestral homologous chromosome pair . During mammalian evolution , these chromosomes lost homology due to progressive degradation of the Y chromosome . The decay of the Y chromosome started with the introduction of sex determination factors , which initiated subsequent cycles of suppressed recombination [1] . Two main mechanisms are usually invoked to explain the reduction of XY homology . Reduced recombination rates near the pseudoautosomal boundary ( PAB ) would result in an accumulation of mutations which ultimately result in the inability to recombine [2] . Suppressed recombination led to the gradual decline in recombination . In addition , a stepwise reduction of recombination has been observed in the mammalian Y chromosome . Based on the nucleotide divergence between the human X and Y chromosome , nine different regions , termed strata , can be distinguished [3] , [4] . It has been speculated that chromosomal rearrangements , such as inversions , might explain the stepwise decrease in sequence similarity between genes ordered on the human X chromosome and their homologs ( called “gametologs” ) on the Y chromosome [3] . Nevertheless , comparative genomic analysis has failed to identify such inversions [5] , [6] . Hence , the forces driving recombination suppression remain to be established . The observation of a gradual demise of the Y chromosome has lead to speculations that , from an evolutionary perspective , the Y chromosome is doomed to extinction [7] , [8] . In contrast , recent evidence suggests that gene loss has been limited over the past 25 million years [9] . However , the Y chromosome is not only shaped by loss of genes and gene functions , but also by addition of genes as a result of interchromosomal exchanges . An autosome to gonosome translocation occurred after the divergence of the placental mammals from marsupials , increasing the size of the eutherian gonosomes [10] . Similarly , chromosomal regions known to be autosomal in different mammals seem to have translocated to the subtelomeric region of the long arm of the X chromosome during great ape evolution and subsequently to both the X and Y chromosome subtelomeres during hominid evolution [11] , [12] . There remain two regions of homology: the 2 . 7 Mb pseudoautosomal region 1 ( PAR1 ) at the telomeres of the short arms and the 0 . 3 Mb human specific PAR2 at the termini of the long arms [2] , [13] . These XY homologous regions are required for pairing and synapse formation resulting in the obligate cross-over which is required for proper chromosome segregation during mammalian meiosis . Although often perceived as stable , it is well known that genes in the human PAR1 show elevated divergence with their primate orthologs and high levels of structural polymorphism . [14]–[17] . Relative to other mammalian species with a characterized PAR , the human PAB ( i . e . PAB1; the human specific PAR2 has no counterpart in other genomes ) is positioned distally . The PAB maps within the gene coding for one of the XG blood group antigens [18] , [19] . XG is disrupted on the Y chromosome , and thus lacks nine exons on its 3′ end . The PAB was probably created by the intrachromosomal transposition of a chromosome fragment including the sex-determining region Y ( SRY ) gene [20] . Over time the PAB has shifted about 240 bp into the PAR by attrition , accounting for the fact that the PAR is flanked by a 240 bp segment of reduced homology ( ∼77% ) on its proximal side [21] . An Alu element is located at the human PAB , but this is believed to have inserted after the divergence of old world monkeys and great apes and therefore did not create the PAB [22] . Hence , the PAB has remained stable since the divergence of the great apes and is considered stable in the Catarrhini lineage [1] , [22] . Here we demonstrate that a previously undiscovered PAR1 length polymorphism exists in the human population as a result of recent recurrent chromosomal rearrangements that shifts the PAB by 110 kb towards the centromere .
To identify pathogenic copy number variants in patients with developmental disorders , we screened ∼4300 individuals ( ∼60% male ) by microarrays . This screening identified 15 male patients of mainly Belgian origin carrying a duplication with a minimum size of 98 , 630 bp and of maximum 136 , 609 bp on Xp22 . 33 ( Figure 1 A ) . To determine whether the duplication occurred de novo or was inherited , arrays were performed on both parents in all 6 families for which parental blood samples could be obtained . We had assumed that the duplicon would have arisen de novo or would be inherited from the mother , since males inherit their X chromosome from the mother . However , the duplication was paternally inherited in all families . Based on the paternal inheritance , we hypothesized the duplicon resided on the Y chromosome . To test this hypothesis and to determine the location of the duplicon , we performed FISH on metaphases from one index , his father , and a male control using probes to target PAR1 and the duplicated region . We observed PAR1 signals on both X and Y chromosomes in all samples ( Figure 1 C ) . The duplicated region was found on the X chromosome in the index , father , and control , but was also found on the Y chromosome of the index and father ( Figure 1 C ) . The FISH analysis demonstrates that the X duplicon is actually located on the Y chromosome at or near the short arm pseudoautosomal region . Since the duplicated region appeared to be adjacent to the X-linked PAR1 , we reasoned that the Y chromosome copy would also be adjacent to the Y-linked PAR1 . If so , the regular Y-PAR1 boundary would be disrupted in carriers and a PCR spanning the Y-PAR1 boundary should result in an amplification product in controls , but not in duplication carriers ( Figure S1 ) . As expected , the Y-PAR1 boundary specific PCR resulted in an amplification product in normal males , but not in females . In contrast to our hypothesis , the same amplicon was observed in carrier males ( Figure S1 . ) . Hence , the duplicon is not a mere extension of Y-PAR1 . To elucidate the exact location of the duplicon on the Y chromosome we performed targeted capture using a BAC spanning the duplicon as a bait , followed by Illumina sequencing for patient P1 . The capture resulted in a 1228 . 4 fold enrichment of the insert region and generated on average 348 , 529 reads over the bait . Since the BAC spans the duplicon , it was expected that some paired-ends would map back to different locations in the Y chromosome reference sequence , and that some reads would feature split sequences . Unexpectedly , no chimeric pairs or split reads could be detected . Upon closer scrutiny of the aligned reads , the PAB region featured three different types of reads: reference Y-PAB reads , reference X-PAB reads , and SNP containing X-PAB reads . The PAB also showed three stretches of heterozygous SNPs: a 33% allele frequency region flanked by two 50% allele frequency regions ( Figure 1 D ) . Since males should have no heterozygous SNPs in their hemizygous X-specific sequences and pseudoautosomal SNPs should show allele ratios of 50% we hypothesized that one portion of the pseudoautosomal region was duplicated . Indeed , the 33% allele frequency region and the proximal 50% allele frequency region represent a duplication event while the distal 50% allele frequency region represents the normal pseudoautosomal SNPs . The breakpoint was delineated by selecting the most proximal ( chrX:2 , 694 , 303 ) SNP with an allele frequency of 50% and the most distal ( chrX:2 , 694 , 429 ) SNP with an allele frequency of 30% . Interestingly , those SNPs both lie in a long terminal repeat , LTR6B , chrX:2 , 694 , 151-2 , 694 , 702 ( 551bp ) . The most proximal 50% allele frequency SNP was also near a second LTR6B repeat , chrX:2 , 808 , 549-2 , 809 , 097 ( 548 bp ) . These repeats explain that more than 99% of the reads mapping around the SNP frequency changes feature a mapping quality of 0 , an indication of reads that map to multiple locations . The presence of these repeats at both sides of the duplicon also explain the absence of chimeric pairs and split reads . The duplicon thus comprises 110 kb of X-specific sequences as well as 5 kb proximal PAR1 , resulting in the construct illustrated in Figure 2 A . The 5 kb sequence is present as three copies in the patient ( once on his normal X chromosome and twice on his Y chromosome with the X insertion ) , which explains the 33% SNP ratio profile as well as the presence of three different kinds of reads at the PAB . To verify the boundary was at the LTR6B in the index patients and their fathers , PCR was performed using one primer located in the distal PAR1 and another located in the X-specific region distal to the duplicon boundary and predicted to span the LTR6B ( Figure 2 A , green primers ) . Not only are these primers separated by 115 kb in the reference genome , but they also have opposite orientations . As expected , the PCR generated an amplicon in carrier males , but not in male or female controls ( Figure 2 B ) . To confirm the presence of the LTR6B in the amplicon , the PCR products were Sanger sequenced , confirming they contained respectively a PAR1 specific fragment , LTR6B , and an X specific sequence ( Figure 2 C ) . These results demonstrate that the duplicon is an insertional translocation from X to Y that is flanked by LTR6B . During the initial screening for pathogenic copy number variants we also identified a patient carrying a reciprocal deletion ( Figure 3 A ) . Familial analysis showed both the female index as well as the father and sister to be carriers . We hypothesized the deletion to be reciprocal to the duplication and to have occurred by non-allelic homologous recombination ( NAHR ) between the LTR6Bs , thus comprising chrX:2694151-2808549 ( i . e . from one to the other LTR6B ) . To confirm this hypothesis we designed a PCR using one primer distal and one proximal of this range . The primer sites are 115 kb apart so the PCR should fail in any individual except those featuring a deletion between them . As expected , the PCR generated an amplicon in the three carriers , but not in male or female controls ( Figure 3 B ) . Therefore , we concluded that the deletion corresponds to the duplication described above . This was verified by Sanger sequencing the index carriers amplicon . The sequence showed an LTR6B featuring an X-specific profile at the region-specific LTR6B positions 2 , 10 , and 15 , and a PAR1-specific profile at positions 3 to 9 and 11 to 14 . Position 1 was not sequenced ( Figure 3 C ) . A G was sequenced at rs2534626/2316283 . We considered this confirmation that a deletion of chrX:2694151-2808549 results in a merged LTR6B . The Y-PAR1 extension observed in the different families could have arisen as a single ancient insertional translocation event or might have occurred recurrently . The finding of a reciprocal deletion supports the latter , but to provide additional evidence the relatedness of the Y chromosomes containing the Y-PAR1 extension was determined by chromosome Y SNP/STR typing . All six father Y haplotypes were identical to their index sons . The carriers featured two main Y-chromosomal haplogroups: all 20 carriers of Belgian origin featured haplogroup I ( sub-haplogroup I2a*; I-P37 . 2* ) and the two French carriers featured haplogroup R ( sub-haplogroup R1b1b2a1a2*; R-P312* ) ( Figure S2 ) . A comparative Y-STR network analysis with other autochthonously Belgian I2 ( I-P215 ) samples revealed that all but one of the I2a* haplotypes displaying an elongated PAR1 belonged to two closely related clusters ( Figure 4 ) . While these two clusters were related to the other I2a* samples earlier observed in the Belgian population the highly distinct I2a*-haplotype appeared to be more closely related to haplogroup I2b1* ( I-M223* ) . Within each of the two clusters of I2a* haplotypes a high paternal kinship was observed as most of their haplotypes differed from each other in less than eight Y-STR loci . According to the mutation rates measured by Ballantyne et al . [23] and the formulae of Walsh [24] the latest common patrilineal ancestors of the largest cluster lived between 7 and 33 generations ago ( 95% credibility interval ) , i . e . between 1185 and 1835 ( generation span of 25 years ) or 855 and 1765 ( generation span of 35 years ) . To determine whether more individuals with the Y-chromosomal sub-haplogroup I2a* carry the described duplicon , we analyzed two additional sub-haplogroup I2a* Belgian individuals ( identified as § in Figure 4 ) . Interestingly the Y-extended PAR1 boundary PCR described above also generated amplicons in these individuals . We therefore conclude that the majority of – if not all – I2a* Y chromosomes carry the duplicon . Formal proof that the extended PAR sequence is truly pseudoautosomal would be the demonstration of recombination events within the duplicon between X and Y . We show two lines of circumstantial evidence that this is the case . First , we amplified ∼5 kb of the distal duplicated region and sequenced the amplicons using long-read single molecule sequencing by PacBio . Haplotypes were phased and all family members were found to share one of three X specific haplotypes ( Figure 2 D ) . All fathers and sons shared at least one similar X specific haplotype , confirming the paternally inherited haplotype . Notably , only family 4 did not have paternal inheritance from the large haplotype groups 1 or 2 . This family was also the only family with haplogroup R from Y-Chromosomal STR typing , and was the only non-Belgian family , confirming its distant and independent origin . Oddly , two samples ( P10 and P15 ) , did not have a single unphased contig ( indicating two identical amplicons ) or a single contig with phasing ( indicating the same target amplicon sequence with small nucleotide differences ) . They instead featured one unphased contig resembling the targeted X region ( P10 56xcoverage , P15 81xcoverage ) and one contig with less coverage ( P10 38xcoverage , P15 32xcoverage ) matching a highly similar ( 95% ) region of the Y chromosome . Despite the observation that all I2a* sub-haplogroup members are derived from an ancient NAHR recombination event , there are at least two different extended PAR haplotypes on the Y chromosomes ( Table 1 ) . The different haplotypes on Y could have developed through historical mutational events or as a product of recombination between the X chromosome and the duplication insertion on the Y chromosome . Since the two main haplotypes differ by twelve variants we believe recombination is a more likely event . Second , ∼5 kb amplicons spanning the Y-extended PAR junction were PacBio sequenced and the smaller PCR amplicons spanning the boundary were Sanger sequenced . Within the Sanger sequenced region , the reference LTR6B sequences marking the duplicon's borders differ between each other at 15 positions ( Figure 2 C ) . All carrier Sanger sequences had the same distal 13 differences that match the X-specific LTR6B . However , the proximal two differences ( SNP rs2534625/rs12843082 and a single indel ) were only found in the LTR6B sequence of a subset of carriers . Carriers with the two variants also had an additional non-reference LTR6B sequence variation ( rs2316283 ) . We labeled the additional variant containing sequence as Junc1 and the more X-specific sequence as Junc2 . All carriers also showed three variants in the pseudoautosomal reference sequence: an A at rs2534627 , a T at rs2857320 , and at rs34061732 . PacBio amplicon sequencing of the duplicon border identified variants upstream of the Sanger sequencing for rs211654 , rs10625422 , and rs211655 in all samples with longer amplicon lengths ( Table S2 , all samples except P10-13 ) . In the region overlapping the Sanger sequences , all PacBio amplicons showed perfect concordance with the Sanger calls . Additionally , a Junc1 carrying an additional SNP ( rs211656 ) was found in a single PacBio amplicon ( termed Junc1c in Table 1 ) . To conclude , within the I2a* sub-haplogroup members , two different Y-extended PAR junctions were identified . Junc1 is a recombination between the distal X specific LTR6B and the proximal PAR specific LTR6B , a consequence of the non-allelic recombination event . In Junc2 , the original X specific LTR6B is observed . The most likely explanation is that a recombination event occurred in the manner of gene conversion with the X-linked LTR6B serving as the donor and the Y-linked LTR6B as the acceptor reconstituting the original X specific LTR6B sequence . Another possibility is that a recombination event occurred within this LTR6B between the distal and proximal parts of the fusion LTR6B reconstituting the original X specific LTR6B sequence .
PAR1 is common to most eutherian mammals , with a gene order that has been fairly well-conserved since its addition to the pre-existing sex chromosomes that are shared with marsupials [25] . However , the PAB has shifted over evolutionary time and both the size and gene content of PAR1 differ among mammalian species , implying genes within the ancestral PAR have been differentially subsumed into the non-recombining regions in different mammalian lineages [2] . In general , there is evolutionary pressure to expand the non-recombining region resulting in contraction of the PAR . This attrition is attributed to recombination suppression of sex determining region flanking regions . This PAR attrition is counteracted by the addition of new PAR sequences via translocations . Such terminal translocation patterns have shaped the human PAR1 and PAR2 [25] . Here we demonstrate that an apparent male specific X chromosomal copy variant flanking PAR1 represents a PAR1 length polymorphism . The duplication causes a 110 kb proximal extension of PAR1 . The PAR1 extension resulted from an insertional translocation event , caused by non-allelic homologous recombination ( NAHR ) between the LTR6B repeats , one of which is located within PAR1 and one in the X-specific region ( Figure 5 ) . This PAB length polymorphism could reflect either an ancient PAB which has shifted during hominoid evolution or could be a de novo event which has occurred during recent human evolution . Several lines of evidence suggest the latter . First , the PAB in great apes and macaques coincides with the human reference PAB . Hence , for this PAB polymorphism to be ancient , it should have arisen during hominoid evolution and subsequently be lost in the majority of our population . Second , the analysis of chromosome Y haplotypes carrying this duplicon shows the presence of this rare variant in different haplogroups which are phylogenetically unrelated ( Figure S2 ) and from different geographic locations , together with an absence of the rare variant in other haplogroups . To be the PAB predecessor , it would have to be lost several times during human evolution . Hence , the most parsimonious explanation is that the PAB polymorphism is caused by independent , but identical , insertional translocation events . In addition , the identification of one family carrying the reciprocal microdeletion which results in a 5 kb reduction of PAR1 on the X chromosome demonstrates that NAHR at the X and Y LTR6B repeats is recurrent . We not only show sequence homology of the extended PAR region , but provide strong evidence for recombination . This provides formal proof that the extended PAR also acts functionally as a pseudoautosomal sequence . Since all the haplogroup I2a* individuals are ancestrally related , the PAR1 extension is likely to be the consequence of a single insertional translocation event . Nevertheless , different X haplotypes exist in the Y of this haplogroup I2a* . We therefore assume that those haplotypes have been introduced by X-Y recombination . Second , three ( four if including Junc1c ) of those males have a different junction ( Junc2 ) as compared to the majority of this haplogroup . We deduced that Junc1 is generated by the insertional translocation event and that Junc2 has arisen by X-Y recombination within the X specific LTR6B . Insertional translocation events are thought to be the consequence of at least three chromosomal breaks resulting in an interchromosomal transposition of a broken fragment ( 2 breaks ) into another chromosome ( at least one break ) . The exact mechanism by which insertional translocations are generated remains , however , to be established . Interestingly , Durkin , et al . [26] showed that several insertional translocation events in cattle genome evolution have occurred via a circular intermediate which subsequently integrated into the receptor chromosome . Here , we present , to our knowledge , the first example of an insertional translocation which is generated by NAHR . Whereas the duplicon is technically an insertional translocation , it has mechanistically arisen because of a terminal translocation event between the long terminal repeat ( LTR ) in unique X chromosomal sequence and the LTR in the XY homologous region . Several reciprocal balanced and unbalanced translocations have been shown to be the consequence of NAHR between different chromosomes [27] . This translocation event can be considered mechanistically similar . In contrast to the known low copy repeats ( LCRs ) that generate NAHR events , the LTRs here are extremely short , with only 548 bp homology . Hence , opposite to the general view that only LCRs larger than several kb are drivers of genomic disorders [28] , [29] , this observation provides further proof that short repeats also have to be considered as drivers of illegitimate recombination [30] , [31] . Whether NAHR is a common mechanism for the generation of insertional translocations remains to be determined . Nevertheless , it is tempting to speculate that the proximal PAB expansion detected amongst mouse subspecies also occurred as a consequence of an interstitial NAHR mediated translocation event . The house mouse , Mus musculus domesticus , has the smallest PAR amongst those that have been mapped . The PAB is located at about 700 kb from the distal end of the X chromosome , the third intron of the Mid1 gene and truncates the 5′ end of the Y copy . However , in Mus musculus castaneus , a subspecies of the house mouse , the PAB shows a 430 kb shift proximal of the M . m . domesticus boundary . The dichotomy in sequence divergence between the proximal and distal segments of the M . m . castaneus PAR suggests the proximal segment is a recent translocation of X chromosome sequence to the Y chromosome . Interestingly , the Y specific PAB of M . m . castaneus is characterized by a long interspersed nuclear element ( LINE1 ) that is present throughout the mouse genome [32] . Since NAHR between line elements has been suggested as a cause of chromosomal rearrangements [33] , [34] it seems plausible that it originated by the same NAHR mechanism . Sequencing of more PABs in other species and populations will probably reveal more pseudoautosomal boundary polymorphisms . Variation in the PAR boundary is likely to have consequences for the expression of both adjacent genes situated in the duplicon: XG and GYG2 ( Figure 1 B ) . XG encodes a surface protein expressed on red blood cells that belongs to a clinically irrelevant blood group system [35] . GYG2 encodes glycogenin-2 , the predominant glycogenin isoform in the liver , which serves as a primer for glycogen synthase [36] , [37] . Since the duplication is inherited in all families where the inheritance could be determined and since the duplication can be traced within most likely all I2a* sub-haplogroup members , it seems clear that this variant does not cause developmental anomalies or observable adverse fitness effects . Loss of XG and GYG2 may , however , have biochemical consequences and is likely to result in reduced fitness . Based on the paternal origin of the deletion and the apparent normal phenotype of the father , any effect of nullisomy of both those genes is likely to be minor . However , follow up of this family as well as the detection of more patients with this deletion is required to establish potential phenotypic effects . In conclusion , we demonstrate that a pseudoautosomal length polymorphism exists in the human population . The extension of the PAR by NAHR presents a novel mechanism shaping sex chromosomal evolution . It seems plausible that such events have occurred frequently during genome evolution . In addition to the already known deceleration of Y chromosome degradation , our results demonstrate a new way of counteracting the processes leading to a loss of Y chromosomes in humans since a proximally extending PAR1 reconstitutes X-borne genetic material thought to be lost from the Y . Thus , if true , current predictions on when the Y chromosome will be lost from humans [38] need to be adapted for the effects of proximally extending PARs . The finding of this length polymorphism also has consequences for statistical genetic analysis in this region since recombination events may alter the haplotype structure periodically .
The study of variants in the human population is part of the analysis of copy number variants coming from an institutional genome wide analysis study . This has been approved by the institutional review board under protocol nr . S55513 . Blood samples were obtained from 16 patients , 13 parents , and 2 siblings referred for cytogenetic investigation due to the presence of intellectual disability ( ID ) , autism , dysmorphic features or - in one case - subfertility . All families had a geographic origin in Belgium , except family 4 that originated from France . Phenotypes are described in further detail in Table S1 . Duplications were confirmed by FISH using BAC clones as previously described [39] . Probe RP11-457M7 was used to target the pseudoautosomal region and probe RP11-146D5 targeted the X-specific duplication . Samples were analyzed on CytoSure 105K and 180K Custom Microarrays composed of probes from the CytoSure Syndrome Plus v2 array supplemented with probes from the CytoSure ISCA v2 60K array [40] , [41] . DNA digestion , labeling , and hybridization were performed according to the manufacturer's protocol . PCR was used to identify the breakpoints . Primers ( Table S2 ) were designed with Primer3 [42] , [43] . Input sequences were masked for interspersed repeat sequences using the RepeatMasker track [44] , [45] provided by the UCSC browser [46] , [47] . Amplification of fragments was performed using the Platinum Taq DNA Polymerase system ( Invitrogen ) , following the manufacturer's protocol . The thermocycler profile used was: 94°C for 30 sec , followed by 25 cycles at 94°C for 30 sec , 60°C for 30 sec , and 72°C for 2:30 min , with a final extension of 72°C for 1 min . We performed Sanger sequencing of the breakpoint-spanning amplicons on an ABI 3130xl automated capillary DNA sequencer ( Applied Biosystems ) . First , ExoSAP-IT ( USB ) treatment was performed according to the manufacturer's protocol . A BigDye Terminator v3 . 1 Cycle Sequencing Kit ( Applied Biosystems ) was then used as follows: the sequencing reaction was performed using 2 µl template , 1 . 5 µl sequencing buffer ( 5X ) , 4 . 5 µl distilled water , 0 . 5 µl Big Dye , 2 . 5 µl primer ( separate reactions for F and R ) . Reaction conditions were: 3 min at 96°C followed by 25 cycles at 96°C for 10 sec , 5 sec at 50°C , and 4 min at 60°C . Sequencing products were precipitated using 10 µl sequencing product , 10 µl distilled water , 2 µl NaAcEDTA ( 1 . 5 M NaAc + 2 . 5 mM EDTA ) and 80 µl ice cold EtOH ( 100% ) . Samples were stored for 15 min at room temperature ( RT ) , and then centrifuged for 30 min at 4°C and 3000 rpm . Supernatant was removed . Samples were centrifuged upside-down for 1 min at 4°C and 1800 rpm . 150 µl ice cold 70% EtOH was added and samples were centrifuged for 10 min at 4°C and 1800 rpm . Supernatant was removed and samples were centrifuged upside-down for 1 min at 4°C and 1800 rpm . Samples were kept for 30 min at RT ( dust- and light free ) . 15 µl of High Dye ( formamide ) were added before spinning and vortexing samples . Samples were stored for 15 min at RT and denatured for 3 min at 96°C . DNA sequences were visualized using ABI sequence scanner v1 . 0 ( Applied Biosystems ) . BAC-mediated targeted paired-end sequencing was used to narrow down the breakpoint region . DNA was captured by a BAC mediated pull-down using an adapted protocol of Bashiardes et al . [48] . BAC clone ChrX-32k-3P23 was labeled with BioPrime DNA Labeling System ( Invitrogen ) according to the manufacturer's protocol . Genomic DNA was sonicated to a fragment size of approximately 350-650 bp and linkers were added . Separately , 300 ng biotin-labeled BAC mixed with 30 µl Cot-1 DNA , 1% 3M Na-Acetate , and 3000 ng fragmented genomic DNA mixed with 1% 3 M Na-Acetate and in 2 . 5× abs . EtOH were precipitated at −20°C overnight . Samples were centrifuged for 30 min at 4°C and 3000 rpm . Supernatant was removed . Samples were centrifuged upside-down for 1 min at 4°C and 1800 rpm . 150 µl ice cold 70% EtOH were added and samples were centrifuged for 10 min at 4°C and 1800 rpm . Supernatant was removed and samples were centrifuged upside-down for 1 min at 4°C and 1800 rpm . Pellets were dried at 37°C for a few minutes and resuspended in 25 µl nuclease–free water at 37°C for at least 30 min . Samples were transferred to 0 . 2 ml tubes and denatured and hybridized in a thermocycler as follows: BAC DNA was denatured for 5 min at 95°C and incubated for 15 min at 65°C . 24 µl of 2× hybridization buffer ( 1 . 5 M NaCl , 40 mM Na-phosphate buffer pH 7 . 2 , 10 mM EDTA pH8 . 0 , 10× Denhardt's Solution , 0 . 2% SDS ) were added in the cycler and samples were incubated for another hour at 65°C . Genomic DNA was denatured after 40 minutes in another thermocycler for 5 min at 95°C and incubated for 15 min at 65°C . 25 µl of 2× hybridization buffer were added in the cycler . Finally both samples were mixed in the thermocycler by pipetting and hybridized for another 70 h at 65°C . 100 µl magnetic beads ( Dynamed M-280 Streptavidin; Invitrogen ) were washed twice with 1 ml binding buffer ( 10 mM Tris-HCl pH 7 . 5 , 1 mM EDTA ph 8 . 0 , 1 M NaCl ) and resuspended in 150 µl binding buffer . Hybridization solution was pipetted from cycler to the prepared beads . Tubes were sealed and spun for 30 min upside-down at RT . Beads were pulled down with a magnet , the supernatant removed , and beads were washed once for 15 min in 1 ml 1× SSC , 0 . 1% SDS on a vibrating table at RT and twice for 15 min in 1 ml 0 . 1 × SSC 0 . 1% SDS at 65°C in a vibrating heating block . 50 µl 0 . 1 M NaOH were added to the beads and they were shook gently for 10 min at RT . Finally the supernatant was pipetted to 1 M Tris-HCL pH 7 . 5 and the resulting volume of 100 µl was distributed on QIAquick Spin Columns and purified according to the manufacturer's instructions . Samples were sequenced on a HiSeq 2000 ( Illumina ) for 2×100 bp reads using the SBS sequencing kit v3 following the manufacturer's protocol . The standard Illumina primary data analysis work-flow was followed for base calling and quality scoring . Illumina reads were aligned to the human genome ( hg19 ) with BWA v0 . 5 . 9 [49] with default settings except that nucleotides with quality score lower than 15 were soft clipped . Read duplicates were discarded after mapping with PICARD MarkDuplicates v1 . 38 ( http://picard . sourceforge . net ) . Local realignment around indels was performed with RealignerTargetCreator and IndelRealigner from GATK v1 . 0 . 4974 [50]–[52] . Finally base quality scores were recalibrated with CountCovariates and Table Recalibration from GATK . The variant frequency of each position of chromosome X between bases 2 , 680 , 000 and 2 , 830 , 000 was assessed with SNIFER ( E . Souche , personal communication ) . Reads not mapped in proper pair , reads mapped with a mapping quality lower than 30 , and nucleotides with quality lower than 20 were discarded . A call was considered heterozygous if its read depth was of at least 100 and the variant frequency was between 25% and 75% . In total , 42 Y-STR loci were genotyped for all samples as described in previous studies [53] , [54] . However , instead of PowerPlex Y the recent developed PowerPlex Y23 System ( Promega Corporation , Madison , WI , USA ) was used . All 42-Y-STR haplotypes were submitted to Whit Atheys Haplogroup Predictor [55] to obtain probabilities for the inferred haplogroups . Based on these results , the samples were assigned to specific Y-SNP assays according to previously published protocols [53] , [56] to confirm the main haplogroup and to assign the sub-haplogroup to the most accurate level of the latest published Y-chromosomal tree [57] , [58] . GenAlEx version 6 . 5 [59] , [60] was used to calculate the differences between all observed haplotypes based on the 42 Y-STR loci . Next , the median joining haplotype network for all the samples belonging to haplogroup I2a* ( I-P37 . 2* ) was constructed based on 26 single-copy Y-STRs by NETWORK version 4 . 5 . 1 . 0 [61] ( http://www . fluxus-engineering . com ) together with all I2 ( I-P215 ) samples already observed in the autochthonous Belgian population by Larmuseau et al . [53] , [54] , [56] . The network analysis used the weighting scheme described by Qamar et al . [62] due to different mutation rates among the markers based on Ballantyne et al . [23] . Primers ( Table S2 ) were designed with Primer3 to cover chrX:2 , 718 , 644-2 , 723 , 016 and a combination of chrX:2805180-2809097 plus chrY:2644703-2645415 . PacBio specific barcodes with padding sequence were added to the 5′ end of the primers . PCRs were performed using the TaKaRa long range PCR kit by ClonTech . Products were checked on agarose gels , individually purified on MinElute columns ( Qiagen ) , quantified with the Quant-iT PicoGreen dsDNA Assay Kit ( Life Technologies ) , and equimolar amounts pooled . This pool was purified on Qiagen MinElute columns , concentrated , and fragmentation checked on a DNA 12000 chip analyzed on a Bioanalyzer 2100 ( Agilent ) . ∼2 µlg of the pool was prepared for sequencing according to Pacific Biosciences 5 kb protocol using PacBio's DNA Template Prep Kit 2 . 0 ( 3 kb-10 kb ) . The library was first sequenced on a PacBio RS using a DNA/Polymerase Binding Kit 2 . 0 on a single SMRT cell for a 120 minute movie . The library was sequenced a second time on a PacBio RSII using a DNA/Polymerase Binding Kit P4 on a single SMRT cell for a 180 minute movie . Both runs used PacBio DNA Sequencing Kit 2 . 0 sequencing reagents . Both SMRTcells were analyzed together using SMRT portal version 2 . 2's RS_Long_Amplicon_Analysis . 1 pipeline with the following non-default settings: minimum sub-read length of 4000 , demultiplexing with paired barcodes , and higher stringency on the barcode filtering ( 30 ) . The assembled amplicon contigs were evaluated by command line BLAST [63] against targeted sequences flanked by 100 N's . Contigs were selected with alignment lengths near to the full target length and a minimum coverage of 30 . These contigs were visualized and aligned with ClustalX 2 . 1 [64] to call variants , ignoring Poly-N length variants of ≥5 nucleotides .
|
The human sex chromosomes differ in sequence , except for homologous sequences at both ends , termed the pseudoautosomal regions ( PAR1 and PAR2 ) . PAR enables the pairing of chromosomes Y and X during meiosis . The PARs are located at the termini of respectively the short and long arms of chromosomes X and Y . The observation of gradual shortening of the Y chromosome over evolutionary time has led to speculations that the Y chromosome is “doomed to extinction . ” However , the Y chromosome has been shaped over evolution not only by the loss of genes , but also by addition of genes as a result of interchromosomal exchanges . In this work , we identified males with a duplication on chromosome Xp22 . 33 of about 136 kb as an incidental finding during a copy number variation screen . We demonstrate that the duplicon is an insertional translocation due to non-allelic homologous recombination from the X to the Y chromosome that is flanked by a long terminal repeat ( LTR6B ) . We show this translocation event has occurred independently multiple times and that the duplicated region recombines with the X chromosome . Therefore , the duplicated region represents an extension of the pseudoautosomal region , representing a novel mechanism shaping sex chromosomal evolution in humans .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"clinical",
"genetics"
] |
2014
|
Pseudoautosomal Region 1 Length Polymorphism in the Human Population
|
The voltage sensitivity of voltage-gated cation channels is primarily attributed to conformational changes of a four transmembrane segment voltage-sensing domain , conserved across many levels of biological complexity . We have identified a remarkable point mutation that confers significant voltage dependence to Kir6 . 2 , a ligand-gated channel that lacks any canonical voltage-sensing domain . Similar to voltage-dependent Kv channels , the Kir6 . 2[L157E] mutant exhibits time-dependent activation upon membrane depolarization , resulting in an outwardly rectifying current-voltage relationship . This voltage dependence is convergent with the intrinsic ligand-dependent gating mechanisms of Kir6 . 2 , since increasing the membrane PIP2 content saturates Po and eliminates voltage dependence , whereas voltage activation is more dramatic when channel Po is reduced by application of ATP or poly-lysine . These experiments thus demonstrate an inherent voltage dependence of gating in a “ligand-gated” K+ channel , and thereby provide a new view of voltage-dependent gating mechanisms in ion channels . Most interestingly , the voltage- and ligand-dependent gating of Kir6 . 2[L157E] is highly sensitive to intracellular [K+] , indicating an interaction between ion permeation and gating . While these two key features of channel function are classically dealt with separately , the results provide a framework for understanding their interaction , which is likely to be a general , if latent , feature of the superfamily of cation channels .
While the entire complement of ion channels in a given cell contributes to the membrane voltage , only a subset ( the voltage-gated cation channel family ) responds significantly to changes in membrane voltage , and the molecular mechanisms underlying their voltage dependence remain the subject of considerable scrutiny [1]–[6] . Voltage-gated cation channels are typified by a modular 6-transmembrane segment ( S1–S6 ) architecture , with the S5 and S6 helices forming a core pore-forming module , and the S1–S4 helices forming a canonical voltage-sensing domain ( VSD ) [7] . The VSD , and particularly a subset of positively charged amino acids in the S4 transmembrane segment , is essential for this voltage-dependent gating [8]–[10] . Inwardly rectifying potassium ( Kir ) channels possess a similar core K+-selective pore module but lack the VSD , and the gating mechanisms of this channel family are generally considered independent of voltage [11]–[13] . Instead , Kir channels are physiologically regulated by ligands specific to each channel subfamily , such as Gβγ subunits ( Kir3 channels ) , protons ( ROMK1 and others ) , or nucleotides ( Kir6 channels ) [14]–[16] . In addition , anionic “signaling” phospholipids such as PIP2 interact with the cytoplasmic domains of all known Kir channels and increase channel activity [17] . Despite clear distinctions at the level of primary sequence , predictions of functional behavior based on structural properties do not always hold firm . For example , CNG channels contain a VSD but exhibit little intrinsic voltage-dependent gating [18] . A similar lack of voltage dependence is apparent in the voltage-sensor equipped KCNQ1 channel when assembled with certain accessory subunits ( e . g . , MiRP1 and 2 ) [19] . On the opposite end of this spectrum , KcsA channels , now an archetypal model for K+-selective pores , appear to exhibit some intrinsic voltage dependence despite lacking a canonical VSD [20] , [21] . A second important uncertainty arises in the mechanism of coupling between the voltage sensor and channel pore . In classical models of voltage-dependent gating ( such as Shaker or other Kv channels ) , the VSD strongly influences opening/closing of the pore-forming domain , in the sense that channel open probability ( Po ) can be reduced to virtually 0 at sufficiently negative voltages and increased to near 1 upon depolarization [22] . In contrast , certain voltage-sensor equipped TRP channels exhibit sustained measurable open probability even at very negative voltages , together with much weaker apparent voltage dependence of gating relative to Kv channels [23]–[25] , and incomplete closure can be engineered in classical Kv channels with open state stabilizing mutations at the S6 helix bundle crossing [26] . Such observations indicate that a model of “tight coupling” between the VSD and pore does not apply to all channel types and that the pore domain itself may strongly influence open probability in some ion channels ( whether equipped with a voltage sensor or not ) . In this regard , voltage-sensitive dynamics of the pore-forming module may not always be obvious in ion channels that are strongly governed by motions of the voltage sensor . Through ongoing characterization of the Kir6 . 2 channel , we have begun to recognize that substitution of charged residues at pore-lining positions can affect channel gating in very unexpected ways . Kir6 . 2 is a two transmembrane domain inwardly rectifying K channel , clearly falling into the realm of “voltage sensor-less” ion channels , and assembles with sulfonylurea receptor subunits ( SUR1 , SUR2A , or SUR2B ) to form KATP channels [27]–[29] . To date , most characterization of KATP gating has focused on its recognized physiological ligands ( notably intracellular nucleotides and anionic phospholipids ) [30]–[32] . The present study reveals remarkable voltage-dependent properties that arise in this “voltage-sensor-less” KATP channel , together with other unrecognized mechanisms of KATP channel regulation by intracellular ions . We have characterized a mutant Kir6 . 2 channel that exhibits marked voltage-dependent gating upon membrane depolarization . The voltage dependence of gating of Kir6 . 2[L157E] is convergent with ligand-dependent gating by ATP and PIP2 and is likely to involve the same “gate” as these intrinsic physiological ligands of the KATP complex . We demonstrate that the voltage- and ligand-dependent gating of these channels is significantly affected by intracellular potassium ions , indicating an interaction between ion permeation and gating and providing a framework for understanding for what is likely to be a general feature of the superfamily of cation channels .
We have characterized the properties of a number of Kir6 . 2 mutant channels substituted with various charged residues at pore-lining positions . Very unexpectedly , we observed that a single point mutation in the pore-forming subunit of KATP ( Kir6 . 2[L157E] ) generates channels that exhibit voltage-dependent activation ( two different patches are depicted in Figure 1A ) . At negative voltages , patches exhibit a steady-state non-deactivating current . Depolarizing voltage steps result in an instantaneous current jump followed by subsequent activation of current , resulting in an outwardly rectifying current–voltage relationship ( Figure 1E ) . These observations contrast with behavior of WT Kir6 . 2 channels ( Figure 1B ) , in which significant time-dependent activation is not typical and the macroscopic current-voltage relationship is nearly linear ( Figure 1E ) . Residue 157 is located at a deep pore-lining position in the Kir6 . 2 inner cavity , directly adjacent to the putative “glycine hinge” ( Figure 1C , D ) . While this single amino acid substitution introduces time-dependent activation somewhat similar to voltage-gated cation channels , the lack of a canonical VSD and a weaker voltage dependence relative to classical Kv channels implies a fundamentally different mechanism is at work ( Figure 1D ) . The effects of glutamate substitution at residue 157 are position specific . We have examined glutamate substitution at multiple other pore-lining positions in Kir6 . 2 [33] and found no evidence of similar behavior in 129E , 160E , or 164E channels ( see Figure S1 ) . Glutamate substitution at position 168 results in somewhat unusual effects on conduction , including intrinsic inward rectification ( in the absence of intracellular blockers ) , but these do not resemble the unique voltage-dependent activation of Kir6 . 2[L157E] . Several observations confirm that voltage-dependent activation of Kir6 . 2[L157E] is due to channel gating , rather than an alternative mechanism such as relief of block , a voltage-dependent change in conductance , or activation of an alternative channel type in the patch . Firstly , we examined the effects of ligands known to alter channel Po in WT Kir6 . 2 and other Kirs , namely PIP2 ( which is stimulatory and enhances open state stability/open probability ) and poly-lysine ( which is inhibitory and reduces open state stability ) . After inside-out patch excision , voltage-dependent currents were inhibited by internal ATP ( Figure 2A , B ) indicating that currents were indeed carried by KATP channels . We subsequently applied either PIP2 or poly-lysine to the cytoplasmic face of the membrane and subjected patches to a series of voltage steps . A pattern emerged in which PIP2 application resulted in accelerated kinetics of activation and a reduction in the activating fraction of macroscopic currents . This effect could be saturated with sufficient PIP2 application , to the point where an activating component of current was no longer apparent ( Figure 2C ) . Application of poly-lysine , which reduces open probability of Kir channels by shielding negatively charged headgroups of anionic phospholipids ( e . g . , PIP2 ) [34] , led to opposite effects: slower activation kinetics and an increased activating fraction of current ( Figure 2C , D ) . To further verify that observed currents are indeed carried by Kir6 . 2 , we exploited the fact that the L157E mutation confers strong spermine sensitivity . Application of spermine to excised membrane patches resulted in complete current inhibition at depolarized voltages ( Figure S2 ) , confirming that the observed voltage dependence is intrinsic to these channels and that development of leak does not contribute to current properties observed after subsequent treatment with activating agents such as PIP2 . Manipulation of open state stability over a wide range ( using PIP2 or poly-lysine ) illustrates a relationship between open state stability and the properties of voltage-dependent gating , demonstrated in normalized current records ( Figure 2E ) and in data from multiple patches ( Figure 2F , G ) . Using ATP sensitivity ( fractional inhibition in 10 µM ATP ) as an index of open state stability [30] , [31] , there is a clear relationship between open state stability and both the activation time constant and the fractional activating component of macroscopic current . At low open state stability ( after poly-lysine , low Po , channels very sensitive to ATP ) , the activating fraction is large and the activation kinetics are slow ( Figure 2E , red trace , and Figure 2F , G ) . In contrast , at high open state stability ( after PIP2 , high Po , channels weakly sensitive to ATP ) , the activating component of current decreases , and the activation time constant is accelerated ( Figure 2F , G ) . PIP2 exposures sufficient to saturate open probability virtually eliminate voltage-dependent gating ( because channels are maximally open at all voltages , Figure 2E , green trace ) . The demonstrated relationship between channel Po and voltage-dependent gating , and especially the loss of voltage dependence at saturating open probability , indicates that the gating of Kir6 . 2[L157E] arises primarily from voltage-dependent changes in Po . Also , convergence of the novel voltage-dependent gating mechanism and intrinsic PIP2 regulation suggests that voltage is influencing the ligand-operated ( ATP/PIP2 ) gate of Kir6 . 2 . It is notable that voltage-dependent properties can vary from patch to patch , as ambient lipid levels ( likely PIP2 ) vary ( Figure 1A , B , Figure 2A ) [35] . The relationship between open state stability and voltage-dependent gating is further illustrated in Figure 3 and provides an additional perspective to the effects described in Figure 2 . After excision , open probability was first brought to saturation by application of PIP2 [30] ( Figure 3A , i ) , and then iteratively reduced with brief poly-lysine applications ( Figure 3A , ii–v ) . Currents after each poly-lysine exposure were normalized to the “fully activated” currents ( condition ( i ) ) . Notably , at low open state stability ( low PIP2 levels , e . g . , Figure 3A , v , or Figure 2D ) , currents at negative voltages are small but can increase several-fold upon depolarization . The result is a large activating fraction of outward current ( in normalized traces , Figure 2E ) , although the absolute currents do not reach the same level as observed in higher PIP2 conditions . As open state stability is increased , basal currents at negative voltages are larger , and the fraction of outward current that exhibits time-dependent activation is necessarily smaller . Importantly , the emergence of these patterns are not due to electrostatic effects on permeation , because neither PIP2 or poly-lysine affect the Kir6 . 2 single channel conductance[31] , [34] . The steady-state voltage and PIP2 dependence of activation of Kir6 . 2[L157E] can be reasonably well fit over a wide range of open state stability with a simple allosteric model ( Figure 3B ) . The model describes an open-closed equilibrium ( KCO ) governed by the membrane PIP2 content , with channels able to occupy two different gating tiers distinguished by the KCO equilibrium constant ( the low Po tier has a small KCO , and the high Po tier a higher KCO—indicated by KCO* in Figure 3B ) . For clarity , the KCO·[PIP2] term directly reflects what we have referred to as “open state stability” thus far . The partition between high and low Po tiers is described by a voltage-dependent equilibrium constant ( Kv ) . The model was fit simultaneously to data over a wide range of open state stability ( by varying [PIP2] in the model ) . Similar experiments and analysis in four patches indicates a Kv ( 0 mV ) equilibrium constant of 0 . 7±0 . 2 ( with an effective valence of 0 . 7±0 . 1 ) and a ∼7-fold stabilization of the KCO equilibrium constant in the high Po tier ( KCO*/KCO = 7 . 5±0 . 9 , this is also the value of the reversibility factor g ) . Although a potential physical mechanism underlying voltage-dependent activation will be discussed in detail in subsequent sections , two elements of this kinetic model are worth noting . In simple terms , the model implies that the Kir6 . 2[L157E] channel operates in two gating tiers ( high Po and low Po ) , with the partition between gating tiers influenced by voltage . Secondly , experimental data seem to preclude any simple model in which an open-closed equilibrium is directly controlled by voltage—such models predict that sufficiently high voltages would open channels to a similar level ( and sufficiently negative voltages would close channels ) , irrespective of basal open probability , a prediction that fails to match the observed behavior ( Figure 3 ) . We also measured currents from patches expressing small numbers of channels ( 1–5 per patch ) to determine the effects of voltage on unitary conductance and Po ( Figure 4A ) . WT Kir6 . 2 and L157E channels exhibit similar single channel current magnitude , indicating that the L157E mutation has little effect on ion permeation ( Figure 4B ) . Consistent with previous reports [36] , [37] , single channel current-voltage relationships also exhibited mild inward rectification . Notably , L157E ( but not WT ) channels exhibit obvious increases in open probability at depolarized voltages ( Figure 4A , C ) . Basal open probability tended to be fairly low in both WT and 157E patches , and so significant increases in open probability were frequently observed for L157E ( Figure 4C , accounting for the large “activating fraction” observed in macropatch records—Figure 2E , Figure 3A , v ) . An important concept of this model is that voltage does not directly drive the channel to open . Rather , channels open stochastically , and rearrangement of ion occupancy after channel opening governs the partition between high and low Po gating modes/tiers . Thus , if anything , the permeant ions themselves can be considered the “voltage sensors . ” If this represents the predominant sequence of events during voltage-dependent activation , then the observed gating kinetics should depend primarily on intrinsic channel opening and closing rates ( rather than the voltage-driven rate ) , and ATP stabilization of the closed state ( prolongation of single channel interburst intervals ) should affect the kinetics of channel opening . This behavior is indeed observed , and the effects can be quite dramatic ( Figure 7A–E ) . Activation kinetics are slowed significantly in 10 µM ATP ( Figure 7B , D ) . In some patches with sufficient current expression and appropriate open state stability , extremely slow activation was also observed in 100 µM ATP ( Figure 7C , D ) . This reflects the infrequency of opening in 100 µM ATP—since openings occur rarely , channels will enter the high Po tier very slowly . These effects can be rationalized by an extension of the simple allosteric model presented earlier ( Figure 7E ) , with the addition of ATP-bound closed states reflecting stabilization of channel closure by ATP . This scheme is not intended to provide a complete description of ATP binding to KATP channels ( see [43]–[45] ) . However , the model describes the important counter-regulation of KATP channels by ATP and PIP2 [30] , [31] and predicts longer channel closures in the presence of ATP . In addition , channel opening upon depolarization and closure after hyperpolarization exhibit very weak voltage dependence ( zact = 0 . 11±0 . 01 , zdeact <0 . 01 , Figure 7F , G ) . Again , this likely reflects the idea that the activation/deactivation kinetics are limited by the intrinsic bursting kinetics of the channel and that voltage is not driving the conformational changes that mediate gating . We have also considered whether intracellular ions might affect channel activity by other mechanisms . We speculated that intracellular ionic strength might affect channel interactions with PIP2 , and this appears to be a definite possibility . To examine PIP2 interactions , WT Kir6 . 2 channel open probability was “rundown” with a high concentration of Mg2+ , and then exposed to various concentrations of diC8-PIP2 , in both high and low ionic strength conditions ( Figure 8 ) . It is clear that in low ionic strength , channels are activated more completely and at lower diC8-PIP2 concentrations . It appears that ionic strength can indeed alter channel-PIP2 interactions , although it should be recognized that this experiment does not establish whether this is a direct effect of ionic strength , or an allosteric effect arising from the actions of ions within the pore ( i . e . , PIP2 interacts with higher affinity with open channels , and so pore-mediated effects of ions on open probability could indirectly affect channel-PIP2 interaction ) . Importantly , it seems unlikely that channel-PIP2 interactions would be altered by mutations deep in the inner cavity—and thus the distinct properties of 157K versus 157E cannot be accounted for by this phenomenon . Nevertheless , there is a possibility that intracellular ionic strength affects KATP channel activity by multiple mechanisms . The KATP complex is a ligand-gated ion channel , in which diverse cytoplasmic ligands ( most notably ATP , ADP , and PIP2 ) determine open probability [30]–[32] , [46] . Nucleotide gating is a unique feature of the Kir6 subfamily , but PIP2 dependence is common to all members [47] . In the present study , we have uncovered an additional dependence on intracellular cations that confers substantial voltage dependence . Changes in membrane voltage markedly alter the open probability of Kir6 . 2[L157E] channels , as confirmed by single channel and macroscopic current recordings . Saturation of Po by PIP2 ( Figures 2 , 3 ) abolishes voltage-dependent activation , confirming that activation reflects increased channel Po . Strong voltage-dependent gating in the absence of a canonical VSD was unpredicted and is remarkable in at least two respects . Firstly , it demonstrates a mechanism by which permeating ions can influence the gating state of the pore-forming module . Secondly , it is imposed on the intrinsic ligand-dependent gating: the kinetic properties and extent of voltage-dependent activation clearly depend on PIP2 ( Figures 2 , 3 ) and ATP levels ( Figure 7 ) , indicating that voltage is influencing the stability of the native PIP2/ATP-operated gate . Converging lines of functional and crystallographic evidence suggest that ligand gating of Kir channels results from closure at or near the inner helix bundle crossing , as it does in Kv channels ( see Text S1 for a detailed discussion of this point ) [1] , [38] , [41] , [48]–[51] . Our data set is consistent with this model—permeant ions play an important role , but there is no ionic selectivity to the effect , and the critical residue ( 157 ) is located in the M2 helix , rather than in the selectivity filter . The voltage-dependent activation of Kir6 . 2[L157E] likely arises from a state preference for one orientation of permeating ions over another ( specifically , whether the cavity site is occupied is vacant ) . Voltage-dependent ion occupancy , as modeled here ( Figure 6 ) , has been inferred from studies of voltage-dependent relief of TEA block in KcsA channels , in which TEA and K+ interactions have been hypothesized to depend on voltage-dependent changes in ion occupancy profiles [40] , [52] . Although specific interactions with channel gating remain unexamined in KcsA and other channels , it is noteworthy that general features for this mechanism ( the K+ channel pore module , with a cavity ion binding site ) are likely present in all K+ channels , and the general principles could extend to other channel types irrespective of structure/sequence . In Kv channels , Po is strongly controlled by the canonical VSD [22] . However , various channel types exhibit considerable diversity in the apparent strength of coupling between the voltage sensor and pore . As alluded to in the introduction , there is growing recognition of nominally “voltage-gated” channels that show far weaker voltage dependence than close Shaker homologues and exhibit persistent open probability at negative voltages [25] . Such features may indicate that coupling between the voltage sensor and pore is relatively weak and that the pore-forming module can significantly affect open state stability/open probability—indeed mutations in the helix bundle crossing region can result in persistent opening of Kv channels [26] . Furthermore , many voltage-gated channel assemblies , perhaps most notably the KCNQ1/KCNE1 complexes , exhibit activation kinetics that appear to be considerably slower than the kinetics of voltage-sensor equilibration [53] . Similarly , a small voltage dependence is generally attributed to the final concerted opening step of the pore module in widely studied channels like Shaker and BK [22] , [54] , although the mechanism for this voltage dependence is not well understood . Growing recognition of diverse non-canonical mechanisms of voltage sensing , in KcsA [20] , [21] , in the present study , and in a recent report of introduced voltage dependence in CNG channels [55] , suggest important avenues to investigate the role of the pore-forming module in controlling open probability . Finally , while Kir6 . 2[L157E] exhibits an obvious voltage-dependent phenotype , the presence of a negatively charged side chain may not be an absolute requirement , since the same underlying feature is weakly detectable in WT channels . A small hint of this phenomenon is apparent in Figure 1C , and we have included a more marked example in Figure S4 . While not as dramatic as the voltage-dependent activation of Kir6 . 2[L157E] , these features can be quite obvious and are exaggerated in modest inhibitory concentrations of ATP . These observations suggest that other features ( beyond electrostatic interactions of charged side chains and the cavity ion ) can generate some state preference for specific configurations of permeant ions . One potential candidate in K+ channels is stabilization of the cavity ion by the pore helices , which may be more prominent in the closed versus open state [56] , and thus might underlie some energetic preference for one configuration of permeant ions over another in different channel states . We have characterized a unique and unexpected voltage-dependent activation feature of a ligand-gated Kir channel . The voltage dependence arises from voltage-dependent interactions of permeating ions with the same gate as that controlled by gating ligands , providing a unifying interaction between two fundamental processes of gating . The effects of the pore-forming module in regulating the kinetics and properties of voltage-dependent gating tend to be overlooked , since voltage dependence of cation channels is generally attributed to motions of a canonical VSD . However , particularly in cation channels that exhibit relatively weak voltage dependence and persistent conductance at negative voltages , we suggest that the pore-forming module itself may be an important structural element in the regulation of voltage dependence and kinetics of channel gating .
Point mutations were prepared using the Stratagene Quickchange kit , on a background of WT mouse Kir6 . 2 . COSm6 cells were transfected with pCMV6b-Kir6 . 2 ( with mutations as described ) , pECE-SUR1 , and pGFP using the Fugene 6 transfection reagent . Patch-clamp experiments were made at room temperature , using a chamber that allowed rapid solution exchange , or the Dynaflow capillary chip-based platform ( Cellectricon Inc . ) , with DF-16 Pro II chips [57] . Data were typically filtered at 1 kHz , and signals were digitized at 5 kHz and stored directly on computer hard drive using Clampex software ( Axon Inc . ) . The standard pipette ( extracellular ) and bath ( cytoplasmic ) solution used in these experiments had the following composition: 140 mM KCl , 1 mM K-EGTA , 1 mM K-EDTA , 4 mM K2HPO4 , pH 7 . 3 . For 50 mM Kint , 300 mM Kint , and 50 mM Kint + 250 mM Naint solutions , all buffer components were kept at the same concentration , with changes only to the indicated principal solutes ( KCl or NaCl ) . Chemicals were all purchased from Sigma-Aldrich , or FLUKA , with the exception of PIP2 ( phosphatidylinositol 4 , 5-bisphosphate , Avanti ) . Models describing steady-state voltage dependence of activation , and ion occupancy , were generated using the “Q-matrix method” [58] . Matrix Q was constructed such that each element ( i , j ) was equal to the rate constant from state i to state j , and each element ( i , i ) was set to be equal to the negative sum of all other elements in row i . State occupancy at time t was calculated as p ( t ) = p ( 0 ) eQt , where p ( t ) is a row vector containing elements corresponding to occupancy of each state in the model at time t . All tasks required for solving these equations were performed in MathCad 2000 . Parameters describing ion occupancy are replicated from an earlier published model describing ion permeation through KcsA channels , with the exception of a repulsion factor describing the interaction of ions in adjacent binding sites [40] . We reduced the published repulsion factor for the simulations described in Figure 6C , as we found this predicted higher cavity occupancy at extreme negative voltages .
|
Ion channels are proteins that regulate the transfer of ions across the cell membrane . The ions travel via a pore formed by the different subunits that constitute the channels , and this pore can be gated by changes in the electrical field across cell membranes . The canonical mechanism underlying voltage dependence of gating relies upon a widely conserved structural motif called the voltage sensor , which undergoes conformational changes when charged amino acids within the motif respond to voltage and consequently affect the opening of the ion channel pore . In the present study , we have identified a non-canonical mechanism that surprisingly generates voltage-dependent changes in the activity of a ligand-gated ion channel that has no voltage sensor . Our observations suggest that ions flowing through the ion channel pore can significantly affect channel activity , and we suggest that voltage-dependent changes in ion distribution in the “cavity site” of the channel can influence opening and closing of the channel independent of canonical voltage sensors .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry/chemical",
"biology",
"of",
"the",
"cell",
"biophysics/biomacromolecule-ligand",
"interactions",
"biophysics/membrane",
"proteins",
"and",
"energy",
"transduction",
"chemical",
"biology/macromolecular",
"chemistry",
"biochemistry/membrane",
"proteins",
"and",
"energy",
"transduction"
] |
2010
|
Voltage-Dependent Gating in a “Voltage Sensor-Less” Ion Channel
|
Skeletal dysplasias are a common , genetically heterogeneous cause of short stature that can result from disruptions in many cellular processes . We report the identification of the lesion responsible for skeletal dysplasia and male infertility in the spontaneous , recessive mouse mutant chagun . We determined that Poc1a , encoding protein of the centriole 1a , is disrupted by the insertion of a processed Cenpw cDNA , which is flanked by target site duplications , suggestive of a LINE-1 retrotransposon-mediated event . Mutant fibroblasts have impaired cilia formation and multipolar spindles . Male infertility is caused by defective spermatogenesis early in meiosis and progressive germ cell loss . Spermatogonial stem cell transplantation studies revealed that Poc1a is essential for normal function of both Sertoli cells and germ cells . The proliferative zone of the growth plate is small and disorganized because chondrocytes fail to re-align after cell division and undergo increased apoptosis . Poc1a and several other genes associated with centrosome function can affect the skeleton and lead to skeletal dysplasias and primordial dwarfisms . This mouse mutant reveals how centrosome dysfunction contributes to defects in skeletal growth and male infertility .
Normal adult stature in humans is achieved primarily through regulation of long bone growth , which occurs through endochondral ossification [1 , 2] . This process begins with the differentiation of mesenchymal stem cells into chondrocytes in regions of the body where skeletal elements will eventually reside . Hypertrophic differentiation of these chondrocytes directs the vascularization of the forming element , allowing osteoblasts to enter and commence mineralization of the cartilage-based template . Pools of cells at the epiphyses of the long bones retain their cartilage identity as a means to secure the progressive addition of bone matrix throughout the period of skeletal growth . These structures are the epiphyseal growth plates . Tight control of chondrocyte proliferation and terminal hypertrophic differentiation allows new bone tissue to replace terminally differentiated chondrocytes in a spatially and temporally regulated manner , ensuring the proper growth of the skeletal elements and the individual overall . Disruption of these processes can lead to skeletal dysplasias . The growth plate maintains a highly ordered architecture to carry out its function and is divided into three distinct zones: the resting , the proliferative , and the hypertrophic zone [1 , 2] . Perturbation of growth plate organization can result in profound growth defects in mice and humans [3–5] . Resting chondrocytes have a low rate of cell proliferation , and the plane of cell division is not controlled . In contrast , rapidly dividing chondrocytes in the proliferative zone undergo directed cytokinesis orthogonal to the main direction of bone growth , and then intercalate to form distinctive columns of disc-shaped chondrocytes [3 , 6] . Major orchestrators of growth plate architecture include the primary cilium [3] , the WNT-planar cell polarity [6] , TGFβ , BMP , and hedgehog signaling pathways [4] . Primordial dwarfisms are a subset of growth insufficiency disorders that are classified as forms of skeletal dysplasia [7 , 8] . Primordial dwarfism typically involves a proportionate reduction in longitudinal growth that commences early in fetal life [7] . Primordial dwarfisms result in profound reductions in height , and many are associated with head and/or facial dysmorphisms [1] . The primordial dwarfisms include Seckel Syndrome , Microcephalic Osteodysplastic Primordial Dwarfism I-III ( MOPD I-III ) , and Meier-Gorlin Syndrome ( MGS ) [7] . These disorders interfere with processes that are intimately connected to the cell cycle rather than an endocrine disturbance . Common pathways that are disrupted in primordial dwarfism patients include: DNA repair [9–12] , DNA replication [13–16] , centrosome dynamics [12–15 , 17–23] , and splicing of minor , or U12 introns , in mRNA [24 , 25] . We report discovery of the gene responsible for autosomal recessive skeletal dysplasia and male infertility in chagun mice [26] . The mutation is a LINE-1 mediated insertion of a processed Cenpw cDNA into Poc1a , which encodes protein of the centriole 1A . This disruption causes exon skipping , and the mutant protein lacks a highly conserved WD-40 domain necessary for normal spindle pole and cilia formation . POC1A is necessary for normal growth plate architecture , craniofacial development , and spermatogenesis . The mechanism underlying the growth defect is cell death and failure to regulate polarized cell division and intercalation at the growth plate . This animal model clarifies the molecular basis for the clinical features in human patients with POC1A mutations and predicts that male patients will be azoospermic [21 , 22] .
The chagun mutation arose on the DBA/2J strain and was mapped to a 6 Mb region of Chr 9 using recombinant progeny from an F1 x F1 intercross with CAST/EiJ [26] . Initial studies revealed an autosomal recessive disproportionate growth disorder , likely caused by disorganization of the proliferative zone of the growth plate . We used dermestid beetle preparations to characterize the mutant bones ( Fig 1 ) . The chagun mutant skull has a slightly domed and disproportionate shape . The skull and mandible are foreshortened along the length , but the widths are similar to wild type . Incisor and molar teeth appear normal . We used micro-CT to assess bone architecture and mineralization ( Fig 1 ) . Representative cross sections and 3D volumetric renderings of 4-month-old male mice were selected based on medial bone mineral content and are shown with matched display settings . We found no differences in cortical mineralization , but the trabecular bone was thinner in both vertebrae and metaphyseal long bones of the mutants . The chagun femora are obviously short , with undulating cortical bone and lateral third trochanter irregularities . Disorganization at the distal femoral growth plate is apparent , likely leading to the noticeable lack of trabecular bone in the metaphysis . Metaphyseal periosteal shaping appears to be significantly impacted by the mutation , with abnormal anterior-posterior and medial-lateral shaping . Similar findings were observed in the tibia , with decreased tibial length , proximal tibial growth plate abnormalities , and excessive proximal tibial and mid-fibular flaring . Skeletal shortening extends to the axial skeleton , with pronounced loss of height in lumbar vertebrae . Genome-wide exome capture and high throughput sequencing of a chagun genomic DNA sample was conducted ( http://www . broadinstitute . org/mouse-mutant-resequencing ) . No obvious deleterious mutations were uncovered in the mapped critical region using this approach . There were some synonymous changes and a single missense mutation that encoded an amino acid present in normal individuals of other species . Some variants were detected in intronic regions captured with the exons , but none appeared likely to disrupt splicing or create ectopic splice donors or acceptors ( S1 Table ) . We extended our search using regional DNA capture to enrich for all non-repetitive genomic DNA in the most broadly defined 8 . 5 Mb chagun critical region ( D9Mit183-D9Mit212 ) . We captured genomic DNA from an obligate carrier ( cha/+ ) and a known mutant ( cha/cha ) , and obtained high throughput sequence with at least 20X coverage for approximately 90% of the bases sequenced from both samples . Manual sequence curation was carried out to detect insertions , deletions , inversions , etc . An insertion of an L1Md-GF-5 end non-LTR transposable element [27] was detected in an intron of the gene that encodes collagen type VI , alpha 6 ( Col6a6 ) . This element is not reported in 16 mouse reference genomes ( http://www . sanger . ac . uk/resources/mouse/genomes/ ) [28] . PCR amplification and Sanger sequencing of this region in DNA samples from affected , unaffected , and DBA/2J mice indicates that it is present in all three samples . Thus , this transposable element does not cause the mutant phenotype . The difficulty inherent in sequencing and mapping repetitive elements likely led to omission of this element from the reference genomes . Manual sequence curation revealed a steep decrease in sequence coverage near exon 8 of Poc1a , which encodes protein of the centriole 1A ( S1 Fig ) . This is more pronounced in the mutant than the heterozygote . In addition , many captured fragments from this region have mismatched paired end sequencing reads , where one read maps to chromosome 9 and the other to chromosomes 5 , 10 , or 12 . PCR amplification of Poc1a exon 8 and flanking intronic regions produced a product that is ~500 bp larger in genomic DNA from chagun mice than in unaffected animals ( Fig 2 ) . Sanger sequencing of the genomic amplification products indicated that exon 8 is disrupted by the insertion of a nearly full-length transcript of centromere protein W ( Cenpw ) including a 61 nucleotide poly-A tail that is 24 bp downstream from the common polyadenylation sequence AUUAAA ( S1 Fig ) . The total size of the insertion is 495 bp , and it includes a 5’ untranslated region ( UTR ) lacking only seven bp relative to the published transcription start site , all three exons , and a 3’UTR . This insertion is unique to chagun; it is not present in any of the 16 additional strains analyzed . The bona fide Cenpw gene is located on mouse chromosome 10 , and processed Cenpw pseudogenes exist on chromosomes 5 and 12 . The sequence of the Cenpw insert in Poc1a shares 100% identity with the exons of Cenpw on chromosome 10 , and 90% and 84% identity with the pseudogenes on chromosomes 12 and 5 , respectively . Thus , the inserted sequences of Cenpw in exon 8 of Poc1a are derived from a processed transcript of the bona fide Cenpw gene . The insertion does not delete any DNA from Poc1a exon 8 . The exon is interrupted by the insertion of the Cenpw transcript in reverse orientation relative to Poc1a . The point of insertion is flanked by short ( 15 bp ) stretches of identical Poc1a sequence ( 5’-CACCGTTGCCTTTTC-3’ ) . This arrangement is consistent with target-site duplications characteristic of LINE-1 retrotransposon-mediated insertions ( Reviewed in [29] ) , and with the consensus LINE-1 endonuclease ORF2 insertion site sequence ( 5’ TTTTC|A ) , just before the insertion of the Cenpw transcript [30–33] . RNA extracted from tibiae was prepared for RT-PCR analysis of Poc1a transcripts . The chagun cDNA produced a smaller amplification product than that detected in normal tibia ( Fig 2 ) . Sanger sequencing of the mutant RT-PCR product revealed that exon 8 of Poc1a is skipped precisely . No splicing into exon 8 was detected in chagun mutants ( Fig 2 ) . Quantitative RT-PCR was carried out using probes designed to amplify exons 2–3 at the 5’ end of the Poc1a transcript and exon 11–12 at the 3’ end of the transcript . The same level of transcripts were detected in RNA from wild type and chagun mutant tibiae using probes for both the 5’ or 3’ ends of the transcript ( S1 Fig ) . This suggests that exon 8 is cleanly skipped and that the mutant transcript is similar in stability to the wild type . Skipping exon 8 preserves the reading frame of the Poc1a transcript . The mutant transcript is predicted to produce a POC1A protein that lacks 23 amino acids within the seventh and final 40 amino acid WD40 repeat ( Fig 2 ) . Western blots were carried out to detect POC1A protein in tibiae using a polyclonal antibody generated against full-length human POC1A , which is highly conserved ( 89% identity , UniProt ) between human and mouse . A single band of similar intensity was detected in wild type and mutant tibia ( Fig 2 ) . Pan-tubulin immunoreactivity was used to normalize protein preparations . The molecular structure of mouse POC1A was modeled using the WD40 structure predictor algorithm [34] . Molecular modeling suggests that the deletion would destabilize the seven bladed propeller structure characteristic of WD40 repeat domain proteins because the first and seventh WD40 repeats normally interlock to form the propeller , and the mutant protein lacks 23 of 40 amino acids that would comprise the seventh propeller . Additionally , the WDSP algorithm predicts certain amino acids to be hotspots for protein-protein interactions based on their biochemical properties and their location on the top face of the propeller structure [35] . There is a predicted hotspot in WD6 , Asn233 , which normally forms hydrogen bonds with Ala274 , a residue that is deleted in chagun mutants . The loss of this interaction likely disrupts the protein-protein interactions of Asn233 . To validate the causal role of the mutation in Poc1a in the chagun phenotype , we undertook a BAC transgenic rescue experiment [36] . The selected BAC clone RP24-384G5 contains four genes in addition to Poc1a: Twf2 , Tlr9 , Alas1 , and Dusp7 . These genes do not cause problems when overexpressed , nor do their known loss-of-function phenotypes mimic any aspect of chagun mutants ( S2 Table ) . Two of three independent BAC transgene insertion sites in founder mice provided completely penetrant correction of the chagun phenotype and caused no abnormalities in control transgene-positive animals . Transgenic chagun animals exhibit normal growth and testicular development ( Fig 3 ) . Body weight and length , as well as testicular weight , appearance , and histology were normal . This evidence supports our assertion that loss of Poc1a function due to the exon 8 insertion causes the chagun mutant phenotype . We obtained embryonic stem ( ES ) cells from the knockout mouse project in which exons 3–6 of the Poc1a locus were deleted and replaced by a gene trap internal ribosome entry site ( IRES ) LacZ and a neomycin resistance expression cassette ( S2 Fig ) . We used these ES cells to generate mice with a null allele of Poc1a [37] . Heterozygous carriers for the LacZ knock-in were normal , and mice homozygous for this null allele phenocopied the growth insufficiency and hypogonadism of chagun mutants ( Fig 3 ) . We performed immunohistochemistry ( IHC ) using an antibody raised against a 50 amino acid peptide that contains the entire seventh WD40-repeat domain of POC1A . We detected POC1A in the proliferative zone of two-week old wild type ( postnatal day 14; P14 ) tibial growth plates ( Fig 4 ) . The discoid chondrocytes in this zone have robust POC1A staining in the cytoplasm . Twenty-one amino acids of the peptide used to raise antibodies are intact in Poc1acha/cha mutants , and staining is observed in Poc1cha/cha mutant tibia . POC1A is also expressed in seminiferous tubules of the testis , and the immunostaining varies depending on stage of the seminiferous cycle . In wild type mice , POC1A protein is detectable in the cytoplasm of Sertoli cells and as single puncta in spermatozoa and spermatids , which likely represents labeling the centrosome of the spermatids as they begin to form flagella [38–40] . Little POC1A staining is observed in the testes of Poc1cha/cha mutants , suggesting that the mutant protein may be unstable or not retained in this tissue . Poc1a knockout mouse testes do not stain with this antibody , confirming the specificity of the antibody for POC1A ( Fig 4 and S3 Fig ) . The expression of POC1A in wild type tibia is consistent with the observed defects in mutant chondrocyte proliferation and expression in wild type male germ cells and Sertoli cells suggests that one or both cell types could be involved in the infertility of male mutants . The centrosome is the microtubule-organizing center of the cell , and it is responsible for formation of the primary cilium and the mitotic spindle , and organization of the Golgi apparatus [41 , 42] . The centrosome is comprised of a mother and daughter centriole , and POC1A is a centriolar protein . Therefore , we hypothesized that the Poc1acha/cha mutation could affect any or all of these centrosome-dependent organelles . To test this prediction , mouse embryonic fibroblasts ( MEFs ) were isolated independently from three wild type and three mutant embryos and cultured for 24 hours on gelatin-coated coverslips . MEFs were either supplemented with serum to promote cell division and formation of mitotic spindles or serum-starved to facilitate visualization of primary cilia . After the culture period , cells were fixed and immunocytochemistry was conducted . Immunostaining with an antibody against acetylated tubulin , which labels the mitotic spindle and the primary cilium , revealed the effects of the mutation on these two cellular structures . MEFS cultured with serum revealed dividing wild type cells with equal bipolar spindles , as expected ( Fig 5 ) . The dividing Poc1cha/cha cells , however , frequently displayed evidence of centrosome amplification , with three or four spindle poles present in a single cell . Centrosome amplification likely results in aneuploidy and cell death . Immunostaining with a GM130 antibody revealed the presence of Golgi near the nuclei of wild type and Poc1acha/cha mutant MEFs . No obvious differences in Golgi organization were noted . The majority of wild type MEFs cultured without serum had a primary cilium detected by acetylated tubulin immunostaining . Significantly more MEFs from wild type embryos had detectable cilia than Poc1acha/cha mutants , 87% vs . 28% respectively . The lengths of the measureable cilia were similar in wild types and mutants , 3 . 06 μm vs . 2 . 67 μm , respectively . Loss of Poc1a function leads to defects in centrosome-mediated processes , including formation of a bipolar spindle and formation or maintenance of the primary cilium in MEFS . The growth plates of vertebrae and long bones of Poc1acha/cha mice are obviously disorganized [26] . Histology of tibiae from the neonatal period to P21 indicates that the growth plate becomes progressively more disorganized with age . At P0 , the mutant growth plates look normal ( Fig 6 ) , but by P15 , the coin stack structure of the chondrocytes in the proliferative zone begins to deteriorate in mutants , and some mutant chondrocytes have a rounded , rather than flat , appearance . By P21 , Poc1acha/cha tibial growth plates are even more severely disorganized ( S4 Fig ) The subcellular position of the Golgi within the columnar chondrocytes of the proliferative zone can provide information about the polarity of the cell [3 , 43] . The Golgi is typically located to the right or left of the nucleus in chondrocytes , towards the edge of the coin stack column . To test whether the Golgi are positioned normally in Poc1acha/cha chondrocytes , immunohistochemistry was conducted on sections of the tibial growth plate utilizing primary antibodies against the Golgi marker GM130 . The lateral subcellular positioning of the Golgi is evident in both wild type and mutant cells in the proliferative zones ( Fig 6 ) . The mutants have more nuclei that are more rounded , and the mutant cells appear mal-rotated relative to the cellular column . This indicates that the chondrocytes in Poc1acha/cha mice exhibit some polarized Golgi localization , but fail to maintain cell shape or ability to re-align after cell division . The primary cilia of chondrocytes in the columns of the proliferative zone are oriented in the direction of the column—pointing toward the resting or hypertrophic zone [3 , 44 , 45] . We conducted immunohistochemistry of tibia using a primary antibody against acetylated tubulin on sections from P15 wild type and Poc1acha/cha mutant mice to test whether primary cilia were properly oriented in the mutant growth plate . Wild type growth plates displayed normal orientation of the primary cilia within their columns . Primary cilia were detected in the Poc1acha/cha growth plate , and some of them are oriented properly ( Fig 6 ) . Formation of multipolar spindles could lead to premature chondrocyte cell death . Ki67 staining is similar in Poc1acha/cha mice , being largely expressed to the proliferative zone . TUNEL staining of P15 tibial sections demonstrated a substantial increase ( 10 . 3 fold , p = 0 . 002 ) in TUNEL-positive nuclei in the Poc1acha/cha growth plate ( Fig 6 ) . Together , these data indicate that the proliferating chondrocytes in Poc1acha/cha mice fail to maintain their cell shape , fail to intercalate properly after cell division , and undergo a significantly higher rate of cell death . Poc1acha/cha males are infertile . In adult mutant males , the latest stage of spermatogenesis that is detected is the pachytene stage , and hypogonadism ensues with age [26] . However , the time of initial germ cell disruption during postnatal development was not previously determined . To address this , we conducted histological analysis of the testis in normal and Poc1acha/cha mutants from birth to adulthood . Mutants were identified by Poc1a exon 8 genotyping and periodic acid-Schiff ( PAS ) histological staining was conducted on cross-sections of testes from males at postnatal day 1 ( P1 ) , P7 , P14 , P21 , and 12 weeks ( Fig 7 ) . Immediately after birth ( P1 ) , the number of germ cell and Sertoli cell nuclei per tubule and the diameter of seminiferous cords in testis cross-sections from Poc1acha/cha males were indistinguishable from wild type counterparts , indicating that during embryonic development the proliferation and migration of germ cells to the genital ridge to form testis cords occurs normally . At P7 , seminiferous cords from Poc1acha/cha males are still morphologically similar to wild type; however , the number of γH2AX positive cells is reduced in mutant males , indicating loss of spermatocytes at the earliest stages of meiosis ( S5 Fig ) . By P14 the morphology of seminiferous tubule cross-sections from Poc1acha/cha mutants are dramatically different from controls . Progression of germ cell maturation to the pachytene spermatocyte stage is clearly evident in tubules of wild-type mice at P14 . In contrast , only a few spermatocytes are detected in the mutant tubules . At P21 , germ cell maturation is still arrested in testes of Poc1acha/cha mutants , and most of the germline appears to be lost in some of the tubules . The testis of sexually mature 12 week old wild type mice contain all stages of spermatogenesis , including undifferentiated and differentiating spermatogonia , primary and secondary spermatocytes , and round and elongating spermatids . In contrast , very few germ cells are present in testis cross-sections from 12 wk Poc1acha/cha mutants . Although some tubules contain a few leptotene or pachytene spermatocytes , most tubules appear to be devoid of all differentiating germ cells and contain spermatogonia only . None of the tubules in Poc1acha/cha adult mice contain secondary spermatocytes or post-meiotic round/elongating spermatids . Taken together , these observations demonstrate that the Poc1acha/cha testis phenotype is caused by a postnatal defect in spermatogenesis beginning at the pre-leptotene stage of meiosis , rather than impairment of testis cord formation during embryogenesis . Sertoli cells function to nourish the germ cells , phagocytose cytoplasmic remnants during spermatogenesis , and promote the translocation of the developing germ cells from the base of the tubule to the lumen . Impaired spermatogenesis could result from defects in Sertoli cells , germ cells , or both . To assess the relative contributions of Sertoli cells and germ cells , we used immunohistochemistry to detect these cell types , with primary antibodies against SOX9 , PLZF ( promyelocytic leukemia zinc finger protein ) and c-KIT . SOX9 is a marker of Sertoli cells . PLZF marks the undifferentiated spermatogonial population that consists of spermatogonial stem cells ( SSCs ) and progenitor spermatogonia that are primed to undergo differentiation [46 , 47] . The expression of c-KIT indicates that spermatogonia have initiated differentiation [48] . At P14 the number of Sertoli cell nuclei per tubule cross-section is not different in wild type and Poc1acha/cha males ( WT = 23 . 3±2 . 3 cells/tubule vs Poc1acha/cha = 26 . 6±4 . 0 cells/tubule , p = 0 . 52 ) . At 3 months of age , the number of Sertoli cells per tubule is also not diminished in Poc1acha/cha testis . The apparent increase in Sertoli cell number per tubule in mutants ( Fig 8A ) is likely due to reduced tubule size and defective spermatogenesis in mutants that arbitrarily causes a greater concentration of Sertoli cells per mm of tubule length . The number of PLZF+ cells per testis section does not differ in wild type and mutant testes at 3 mo ( Fig 8A ) , or at earlier ages ( S6 Fig ) . The quantitation of undifferentiated spermatogonia ( i . e . PLZF+ cells ) in wild type vs Poc1acha/cha testis at each age was as follows: P7: 8 . 1 ±0 . 4 vs 4 . 8±1 . 2 , p = 0 . 07 . P14: 13 . 4±1 . 6 vs . 9 . 5±0 . 2 , p = 0 . 07 . P30: 5 . 9±0 . 9 vs 9 . 5±1 . 4 , p = 0 . 1 . The presence of c-KIT+ spermatogonia indicates that spermatogonial differentiation occurs in both wild type and Poc1acha/cha testes ( Fig 8 ) . The histone marker γH2AX is rapidly phosphorylated in response to double strand breaks , and it recruits DNA damage response factors . It is prominently expressed in all intermediate and B spermatogonia and in pre-leptotene to zygotene spermatocytes . While the proportion of tubules in Poc1acha/cha testes with γH2AX staining is similar to wild types at P7 ( 61%±14% vs 65%±8% , not significant ) , there is a major reduction ( 1 . 9 fold ) in the number of γH2AX-positive cells within these tubules in the Poc1acha/cha mutant testes: 12 . 98% vs 6 . 73% , respectively ( Fig 8 , Panel D ) . The defect is much more severe at P14 ( S5 Fig ) . This suggests a loss of germ cells early in meiosis at the pre-leptotene and zygotene stages , resulting in some spermatogonia at the pachytene stage and none at later stages . Condensation of chromosomes and formation of synaptonemal complexes takes place at these early meiotic stages . Next , we utilized spermatogonial stem cell ( SSC ) transplantation to determine whether the spermatogenesis defect was attributable to poor Sertoli cell function , intrinsic defects of the germ cells , or both [49] . In the first set of experiments , we transplanted Rosa-LacZ marked wild type spermatogonia into the seminiferous tubules of wild type recipients pre-treated with busulfan to eliminate the endogenous germline , or alternatively into adult Poc1acha/cha mutant males with impaired spermatogenesis ( N = 4 recipients and 8 testes per genotype ) . Approximately 3 months after transplantation , X-gal staining revealed that the wild type donor SSCs generated colonies of donor-derived spermatogenesis in wild type recipient tubules , as expected . No colonization was detected in any of the Poc1a mutant testes ( Fig 8B ) . The failure of Poc1a mutant testes to support the engraftment of wild-type SSCs is not attributable to the presence of residual mutant SSCs because donor SSCs can colonize normal testes with normal spermatogenesis occurring , albeit at a much lower rate [50] . These results are consistent with a Sertoli cell defect in the Poc1a mutants that could impair the homing ability of transplanted wild-type SSCs and/or re-establishment of colonies of continual spermatogenesis . Impaired homing seems unlikely to be the major cause because a primary spermatogonial population is established during neonatal development in Poc1acha/cha mutant testes . Thus , migration of germ cells from the lumen of the cords to the basement membrane during postnatal development occurs normally in Poc1acha/cha males . For the second set of experiments , we crossed the Rosa-LacZ marker into the Poc1a cha/+ stock to produce Poc1acha/cha males with LacZ marked germ cells . The marked , mutant spermatogonia were transferred into the seminiferous tubules of busulfan-treated wild type recipient nude mice . As a control , Rosa-LacZ marked wild-type spermatogonia were transferred to identical recipients at the same time . As expected , the wild-type SSCs generated robust colonies of spermatogenesis with uniform blue staining in testes of all recipients ( N = 4 recipient mice and 8 testes ) . In contrast , only small patches of spermatogenesis with non-uniform blue staining were generated from Poc1acha/cha mutant SSCs , consistent with blunted expansion of colonizing SSCs and arrested spermatogenesis . Transplantation of mutant cells did not generate densely stained colonies of donor-derived spermatogenesis in any recipient transplanted testes ( N = 4 recipients and 8 testes ) . Taken together , these findings suggest that the SSCs of Poc1acha/cha mutants are capable of initiating a colonization of a wild type microenvironment , but that intrinsic defects in the germ cells cause arrested spermatogenesis , even with wild type Sertoli cell support . These results are consistent with the idea that POC1A expression in both germ cells and Sertoli cells is important for normal testicular function .
We report the discovery of the genetic basis of skeletal dysplasia and spermatogenic failure in chagun mutants . Insertion of a processed Cenpw transcript into exon 8 of the gene that encodes protein of the centriole 1A ( Poc1a ) causes elimination of a portion of one of the seven highly conserved WD40-repeat domains and abrogates function . We conclude this based on: 1 ) identification of an exonic insertion that causes skipping of exon 8 in Poc1a mRNA transcripts , 2 ) co-segregation of this Poc1a mutation with the chagun mutant phenotype , 3 ) the lack of unique insertions , deletions , or coding region mutations in other genes within the chagun critical interval , 4 ) successful BAC transgenic rescue of the mutant phenotype , 5 ) recapitulation of the mutant phenotype with a null , lacZ knock-in allele , and 6 ) expression of Poc1a in the affected tissues . Taken together , this provides compelling evidence that the insertion in Poc1a causes the chagun phenotype . Failure to detect the insertion by exome sequencing is probably attributable to the fact that all of the exon 8 genomic DNA sequence is still present in the mutants . The insertion of the Cenpw processed transcript increases the size of exon 8 from 69 bp to 564 bp , and increased exon size can cause skipping [51] . Skipping of exon 8 maintains the Poc1a open reading frame , and the predicted POC1A mutant protein lacks 23 amino acids in the most C-terminal , highly conserved WD40 repeat domain . POC1A has seven WD40 repeats that are expected to form a seven bladed , circular beta propeller structure . The lesion in the 7th repeat likely causes a failure of the blades to interlock . The mutant POC1A protein is readily detectable in tibia but not testis of Poc1acha/cha mice , suggesting tissue specific effects on protein stability . The Cenpw cDNA insertion appears to be a LINE-1 mediated event because it is flanked by a 15 bp target site duplication . In addition , the sequence 5’ TTTC|A in wild type exon 8 matches the ORF2 consensus ( reviewed in [29] ) . Cleavage between the C and A at these sites permits target-primed reverse transcription by ORF2 reverse transcriptase [30 , 52 , 53] , resulting in insertion of the transcript into exon 8 of Poc1a by the LINE-1 encoded proteins [54 , 55] . We are aware of only two other examples of LINE-1-mediated mutagenesis that involve non-LINE1 transcripts [56–58] . An insertion of a retrogene into an intron of FGF4 causes short stature in 19 dog breeds , and early dog breeders apparently selected for it . The insertion causes atypical expression of FGF4 in chondrocytes , rather than loss of function that we observed in Poc1acha/cha mice . Chronic granulomatous disease , an immunodeficiency disorder in humans , is caused by a LINE-1 mediated insertion of a partially processed transcript into an intron of the CYBB gene , which encodes a cytochrome that is essential for phagocytosis by leukocytes . The insertion causes loss of function; it disrupts splicing , resulting in incorporation of a novel exon and premature termination . Processed pseudogenes have also been detected in cancer genomes [59] . The difficulty in identifying the chagun mutation by exome sequencing suggests that the paucity of examples of LINE-1 mediated mutagenesis could , in part , due to alignment issues with the programs used to analyze exome sequencing data , resulting in ascertainment bias . Two mutations in POC1A ( p . Arg81X and p . Leu171Pro ) were recently reported in separate consanguineous pedigrees with severe short stature and craniofacial abnormalities [21 , 22] . Sarig , Sprecher , and colleagues labeled the syndrome a primordial dwarfism called SOFT Syndrome ( short stature , onychodysplasia , facial dysmorphism , and hypotrichosis ) [21] . Shaheen , Alkuraya and colleagues reported the p . Leu171Pro mutation , and they noted global developmental delay including cognitive impairment in some affected individuals , but they did not observe hair or nail defects [22] . We observed no fur or nail abnormalities in the chagun mutants or ES derived Poc1a null mutants . There was no discussion of hypogonadism or fertility of the human subjects in either report . The POC1A mutation p . Arg81X permits significant read through translation , which left open the possibility that it is a hypomorphic allele . The growth insufficiency , however , is severe in patients with either mutation . Mice homozygous for the ES-derived null allele of Poc1a have the same growth defect , skeletal dysplasia , and testicular features as Poc1acha/cha mice . This indicates that Poc1acha is a loss of function allele . The variable secondary features in the two families could be due to functional differences between the two alleles , contributions from mutations in other genes , or genetic modifiers . Both mutant mice exhibit severe growth abnormalities and craniofacial dysmorphism , and provide excellent models for the major features of the human syndrome . In metazoans , the centrosome regulates organization of microtubules and progression of the cell cycle [60] . Two centrioles , cylindrical , tubulin-rich structures that , together with additional peri-centriolar material , form a single centrosome in each cell . The centrosome is typically located centrally near the nucleus of the cell , but it moves to the leading edge of polarized , migrating cells . At the G1 to S transition , the centrosome begins to duplicate , and after the G2 to M transition , the mother and daughter centrosomes migrate to opposite poles of the cell and form the mitotic spindles . The centrosomes are not strictly required for spindle formation , but they are believed to enhance its efficacy and ensure the fidelity of cell division . While cells are in the quiescent state the centrosome migrates to the cell surface and forms the basal body of the primary cilium . The centrosome and Golgi apparatus are juxtaposed during interphase and are thought to interact functionally for directional protein transport . Proteome analysis in Chlamydomonas and Tetrahymena identified eighteen different proteins of the centriole , including Poc1 , which is a core component of the centriole and basal body in all organisms with motile cilia [61] . Vertebrates have two genes , Poc1a and Poc1b , which are broadly expressed , exhibit 49% amino acid identity , and have overlapping function in cell culture [62] . Embryonic fibroblasts from Poc1acha/cha embryos exhibit a number of anomalies in centrosome-mediated processes . The majority of dividing mutant fibroblasts form aberrant mitotic spindles , and the primary cilia are infrequent and abnormally shaped . We did not notice any obvious defects in the organization of the Golgi . We observed binuclear Poc1acha/cha MEF cells , which are likely indicative of failed cytokinesis , similar to observations of Poc1 dysfunction in Tetrahymena [63] . Skin fibroblasts from human patients with POC1A dysfunction also had abnormalities in mitotic spindle polarity and the formation and length of the primary cilium [21] , and although centrosome ultrastructure was normal , centrosome number increased and Golgi trafficking was impaired [22] . Knock down of Poc1a in cells leads to an inability to recruit markers of a mature centrosome [62] . Taken together , these data suggest that although POC1A deficient fibroblasts can form normal looking centriolar structures , the recruitment of proteins to the centriole is defective , causing abnormal centrosome function and increased number of centrosomes per cell . This leads to multipolar spindles and failed cytokinesis . The centrosome abnormalities also lead to defects in cilia formation and/or maintenance . Poc1acha/cha chondrocytes likely have defects in centrosomes , spindle poles and cytokinesis similar to those observed in Poc1acha/cha embryonic fibroblasts and POC1A patient fibroblasts [21 , 22] . Cell division in the growth plate is an intriguing process in which chondrocytes undergo polarized cell division orthogonal to the main direction of bone growth and spread back over one another to form columnar structures comprised of disc-shaped chondrocytes [6 , 64] . The growth plates of newborn Poc1acha/cha mutants have normal organization into zones and normal morphology of proliferating chondrocytes , but by puberty the proliferating chondrocytes are losing the typical coin-stack structure , and their shape is abnormally rounded rather than flat . As the proliferating cells become more disorganized , apoptosis is increased , and the final size of the proliferative zone is reduced . Although mutant chondrocytes can produce cilia and exhibit polarization of the Golgi , the function of the centrosomes and cilia are likely impaired . This view is supported by the similar chondrocyte disorganization and abnormally round cells in mice with disruption of Kif3a , a kinesin II motor complex protein required for intraflagellar transport and cilia formation [3] . Cilia defects are known to affect the Wnt/Planar cell polarity pathway , which is important for cellular shape , migration and organization [42 , 65 , 66] . The primary cilium is also important in the cellular response to mechanical cues that govern growth plate structure . For example , disruptions of the osteoblast primary cilium affect mechanoresponsiveness , mesenchymal cell differentiation [67] , and the response to mechanical loading [68 , 69] . Thus , the centrosomal defects in Poc1a mutants could account for abnormal microtubule organization affecting chondrocyte morphology , aneuploidy leading to increased cell death , and poor cilia function influencing cell alignment at the growth plate . Testosterone levels , seminal vesicle development , and Leydig cell structure are all normal in Poc1acha/cha males [26] , suggesting normal function of the hypothalamic-pituitary-gonadal axis . The germline is established normally in Poc1acha/cha males during embryogenesis and early neonatal life , as indicated by normal testicular morphology and PLZF-positive spermatogonia . Hypogonadism and infertility arise later in postnatal development when spermatogonia normally undergo a transition from mitotic to meiotic divisions . Defects in germ cells and Sertoli cells contribute to the infertility . Mutant SSCs transplanted into wild type hosts are capable of initial colonization , but they do not establish colonies typical of normal spermatogenesis . There are examples of spermatocytes with chromosome mispairing or DNA repair defects that are arrested at the pachytene checkpoint and undergo apoptosis [70] . In the current study , we extended the assessment of spermatogenic defects and discovered a reduction of meiotic cells at P7 in Poc1acha/cha males compared to wild type counterparts , which is when pre-leptotene spermatocytes first arise in the male germ cell lineage . Also , we found that the number of spermatocytes is greatly reduced in Poc1acha/cha males at P14 when pachytene stage cells first arise . Together , these findings indicate that defects in germ cell maturation begin to arise at the earliest stages of meiosis but become more pronounced as meiotic progression ensues , eventually leading to elimination of a majority of the population . We predict that Poc1a mutant spermatocytes are aneuploid , which triggers apoptosis at the pachytene checkpoint . Also , our findings suggest that Poc1acha/cha Sertoli cells do not support the engraftment of wild type SSCs or development into mature sperm , possibly due to defects in the microtubule organizing center and/or failure to secrete factor ( s ) essential for spermatogonial survival and/or germ cell maturation . The arrangement of microtubules in Sertoli cells is unique and highly specific to this cell type [71] . Defective centrosomes lacking POC1A could disrupt microtubule organization , leading to defective SSC niches [72] and/or an inability to support cells at later stages of spermatogenesis . The inability of wild-type SSCs to colonize Poc1acha/cha testes suggests that the mutation disrupts the ability of the testis microenvironment to support engraftment of normal SSCs in the niche at the basement membrane . Compromised trafficking of components required for the germ cells , or even the ability of the Sertoli cells to support the development of spermatocytes and spermatids could contribute to impaired spermatogenesis in Poc1acha/cha mutant testes . Immature or multiple centrosomes could impair the ability of Sertoli cells to support differentiation and development of spermatozoa from SSCs . Poc1a and Poc1b are both expressed in testis and bone ( Fig 4 and S7 Fig ) , and have functional overlap in cell culture [62 , 63] . Mutations in human POC1B , however , cause a very different disorder . Homozygotes for the p . Arg106Pro mutation in POC1B have severe , syndromic retinal ciliopathy with defects in kidney and cerebellar function [73] . Individuals with p . Gln67del mutations also have recessive , non-syndromic cone rod dystrophy [74] . The basis for these tissue-specific effects is not clear . A loss of function of Poc1a causes skeletal dysplasia and male infertility in chagun mice . Insertion of a processed cDNA causes exon skipping and predicted deletion of 23 highly conserved amino acids necessary for the structural integrity of the protein . POC1A is expressed in the growth plate of long bones and the seminiferous tubules of the testis , the tissues with the most obvious cellular phenotypes . Several processes contribute to the growth defect . First , proliferating chondrocytes undergo enhanced cell death , likely due to multipolar spindle formation . Second , rapidly proliferating chondrocytes fail to maintain the characteristic flattened cell shape and fail to intercalate into a cellular column after cell division , consistent with defective cilia function . Male infertility is a consequence of the inability of Sertoli cells to support germ cell development , as well as germ cell defects in spermatogenesis . The Poc1acha/cha mouse has provided information on the molecular mechanism underlying clinical features of human patients , and highlights male fertility as a potential area of interest for clinicians examining adult male patients with mutations in Poc1a . The Poc1acha/cha and Poc1atm1 ( KOMP ) Mpb mice are valuable tools for studying the molecular mechanisms of dwarfisms and for testing therapeutic interventions .
The chagun mutation arose spontaneously on the DBA/2J strain in Dr . Linda D . Siracusa’s laboratory at Thomas Jefferson University ( Philadelphia , Pennsylvania ) . The mice were transferred to the University of Michigan ( Ann Arbor , Michigan ) . Recently they have been maintained on a hybrid background consisting of C57BL/6J and DBA/2J background strains , and also outcrossed to the FVB/NJ strain for routine maintenance . The BAC RP24-384G5 was purchased from Children’s Hospital of Oakland Research Institute ( CHORI ) . Transgenic mice carrying this BAC were generated by the University of Michigan Transgenic Animal Model Core [36] . Embryonic stem cells containing the Poc1a tm1 ( KOMP ) Mbp mutation , CSD45930 , were purchased from the KOMP ( University of California , Davis , CA ) . These stem cells , JM8A3 . N1 , were derived from the C57BL/6N-Atm1Brd strain . Two targeted clones were used: DEPD00572_5_B12 ( Poc1a_B12 ) , and DEPD00572_5_C09 ( Poc1a_C9 ) . These stem cells were injected into blastocysts and transferred to pseudopregnant surrogate mothers by the University of Michigan Transgenic Animal Model Core . Chimeras were mated to C57BL/6J females to obtain germline transmission . All mice were housed in a specific pathogen free facility with 12-h light , 12-h dark cycle in ventilated cages with unlimited access to tap water and Purina 5020 chow . All procedures using mice were approved by the University of Michigan Committee on Use and Care of Animals ( UCUCA ) , and the Washington State University Institutional Animal Care and Use Committee ( IACUC ) , and all experiments were conducted in accordance with the principles and procedures outlined in the National Institutes of Health Guidelines of the Care and Use of Experimental Animals . Euthanasia was conducted by CO2 inhalation , except for newborn mice that are unresponsive to this method . Decapitation was used for those animals . Surgery was conducted with approved anesthetics . Experienced veterinary care was provided . Two protocols were used to genotype for chagun . Before the chagun mutation was uncovered , the genotype at cha was inferred based on polymorphic flanking markers . Primers were designed to amplify regions of genomic DNA flanking the chagun critical interval that contained informative SNPs . These SNPs differ between the mutant ( DBA/2J ) and the wild-type ( FVB/NJ ) backgrounds: rs30174769 , rs29637716 . These amplification products were digested with Bsr1 endonuclease and run on a 2% agarose/tris-boric acid-EDTA gel to visualize the different bands that segregated differentially between the two background strains . After the mutation was uncovered , mutants , heterozygous , and wild-type animals were distinguished using primers designed to detect the presence of the insertion from flanking sequence ( two potential products; a small wild-type allele , and the larger chagun allele with the insertion ) . Animals from the BAC transgenic rescue experiment were genotyped using the polymorphic SNP markers , and at least two additional sets of primers that amplify products specific to the BAC backbone . All primers anneal optimally at 60°C . The forward ( Fwd ) and reverse ( Rev ) primer sequences are: Animals were genotyped for the presence of the Poc1a tm1 ( KOMP ) Mbp mutation by PCR with primers for LacZ and primers in Poc1a exons 6 and 7 , which are not present in the Poc1a tm1 ( KOMP ) Mbp mutant allele . Primers for LacZ amplification: Fwd: ATCCTCTGCATGGTCAGGTC Rev: CGTGGCCTGATTCATTCC . Amplification was at 94C for 3 min followed by 35 cycles of 94C 30sec , 58C 30 sec , 72C 30sec , followed by 72C 10min . Primers for detection of wt Poc1a: Fwd: TCTGCTTTGCGGTGTACGAA Rev: TTGGGTAGGGTGGGGTACAT . The conditions for this PCR are 92C for 2 min followed by 30 cycles of 92C 10sec , 57C 30sec , 72C 30sec . Two different capture approaches were taken . The Broad Institute conducted genome-wide exome capture and sequencing ( Mutant Mouse Resequencing Project , The Broad Institute of MIT and Harvard , Cambridge , MA ) . Regional targeted enrichment was performed using custom capture probes targeting non-repetitive sequences within the chagun critical interval , and targeted enrichment libraries were generated ( Roche/Nimblegen , Madison , Wisconsin ) . Genomic DNA samples from both a heterozygote ( known carrier ) and a chagun mutant were used to generate these targeted enrichment libraries , which were indexed separately to allow for multiplexed high-throughput sequencing ( IlluminaHiSeq ) by the University of Michigan DNA Sequencing Core . DNA Sequence reads were aligned to the B6 reference genome ( mm9 ) using BWA [6] . Custom perl scripts were used to identify the location of potential insertional elements , using the paired-end read mapping data , and these were manually reviewed using Broad Institute’s Integrative Genomics Viewer to visualize and evaluate the DNA sequence data [28] . The region around Poc1a exon 8 was amplified by PCR of genomic DNA using the following primer sequences: Fwd: 5’-TGTCCCACTGCCACTGCCACTCA-3’ , and Rev: 5’-GGAAGACTCGCCCCACAGGACTCA-3’ . Optimal annealing temperature: 60C . PCR was conducted with the GoTaq DNA Polymerase and buffers provided by the distributor ( Promega ) . Sanger sequencing was completed by the University of Michigan DNA Sequencing Core . Extraction of total RNA was completed with the RNAqueous 4-PCR Kit ( Ambion ) according to the manufacturer’s instructions . The total RNA was utilized to generate cDNA with Oligo dT Primers ( Invitrogen/Life Technologies ) and Superscript II Reverse Transcriptase ( Invitrogen/Life Technologies ) according to the manufacturer’s instructions . The cDNA was used as a template in standard PCR reactions using GoTaq DNA Polymerase ( Promega ) and the following primers to amplify two regions in the Poc1a cDNA: Optimal annealing temperature: 55°C . Sequencing was completed by the University of Michigan DNA Sequencing Core . TaqMan Universal PCR Master Mix ( Applied Biosystems/Life Technologies ) was used according to the manufacturer’s instructions . The following TaqMan probes were included to test expression levels of Poc1a: Exon 2–3 Assay ID: Mm01235877_m1 . Exon 10–11 Assay ID: Mm01235875_m1 . Reactions were loaded into MicroAmp Optical 96-well reaction plates ( Applied Biosystems/Life Technologies ) and run using an Applied Biosystems 7500 Real-Time PCR System . Isolation of protein from postnatal day 3 tibiae and western blot analyses were carried out as previously described [75] . The blot was incubated with a 1:500 dilution of a mouse anti-human POC1A primary antibody ( Abcam , ab67698 ) overnight at 4°C . The blot was incubated with a 1:5000 dilution of a goat anti-mouse IgG secondary antibody ( Jackson ImmunoResearch Laboratories , Inc . #115-035-003 ) for one hour at room temperature . The blot was stripped and incubated with a 1:5000 dilution of a rat anti-yeast tubulin antibody ( Abcam , ab6160 ) overnight at 4°C . The blot was incubated with a 1:10 , 000 dilution of a goat anti-Rat IgG secondary antibody ( Jackson ImmunoResearch Laboratories , # 112-035-102 ) for one hour at room temperature . All antibodies were diluted in a blocking solution made up of 1% weight:volume Bovine Serum Albumin;Tris-buffered saline with Tween-20 . Adult skulls and were completely stripped of flesh by incubation in a dermastid beetle colony ( http://www . lsa . umich . edu/ummz/mammals/dermestarium/default . asp . ) Briefly , mice were euthanized , and bony parts were trimmed of excess flesh . These specimens were fed to the beetles . After the remaining soft tissues were removed , the bones were frozen at ~-20 C for 72 hrs , and the dead beetles removed . To assess shape and mineralization of bones , mice were euthanized , bones were dissected and skin and flesh trimmed . Specimens were imaged in water using a cone beam microCT system ( eXplore Locus SP; E Healthcare Pre-Clinical Imaging , London , ON , Canada ) . Scan parameters were 0 . 5 degree rotational increment , 4 frames averaged , 80 kVp/80 μA X-ray , and 0 . 508 mm Al filter plus beam flattener to reduce beam hardening artifacts [76] . Volumes were reconstructed at 18 μm isotropic voxel size and calibrated for grayscale value by a manufacturer-provided phantom of air , water , and hydroxyapatite-mimicking material . Tibiae were dissected from 15 day old animals , fixed overnight in 4% paraformaldehyde ( PFA ) , rinsed in PBS , and decalcified in 14% EDTA solution ( weight:volume ) for approximately 7 days , changing the solution each day . Testes were removed from mice of the listed ages and fixed overnight in Bouin’s Fixative Solution ( Sigma ) . Both the testes and tibiae were then dehydrated through an ethanol series and embedded in paraffin . Sections were stained with hematoxylin and eosin ( tibiae ) or periodic acid-Schiff’s reagent ( testes ) according to standard protocols . Immunohistochemistry was conducted after removing paraffin from the samples by incubation of slides in xylenes and rehydration in an ethanol series to 1X PBS . Afterward , the testis sections were boiled in a 100 mM solution of citric acid ( pH 6 . 0 ) for 10 minutes to expose epitopes , and cooled . Tibia sections were fixed for 10 minutes in 4% PFA , washed in PBS , and were either incubated with proteinase K in a buffer containing 50 mM Tris base , 1 mM EDTA and 0 . 5% Triton-X100 ( acetylated tubulin ) , or placed in citric acid ( pH 6 . 0 ) heated to 65°C for 1 hour ( GM130 , POC1A ) . Both testis and tibia sections were incubated for 20 minutes in solution of 3% hydrogen peroxide diluted in methanol to quench endogenous peroxidase activity . The slides were blocked in a solution included in the Tyramide Signal Amplification ( TSA ) Kit ( Perkin-Elmer #SAT701001EA ) or Mouse-On-Mouse ( M . O . M ) Blocking Reagent ( Vector #BMK-2202 ) for one hour at room temperature . This was followed by an overnight incubation with primary antibodies in either TSA Block or Vector Mouse-On-Mouse Diluant at 4C . The rabbit anti-rat POC1A antibody was raised against rat POC1A aa 242–291 , which is homologous to mouse POC1A aa 280–329 . The mouse aa 269–308 comprise the WD7 repeat . The primary antibodies include: rabbit anti-rat POC1A ( Abcam , ab135361 , diluted 1:100 ) , rabbit anti-human PLZF ( Santa Cruz Biotechnology Inc . , sc-22839 diluted 1:50 ) , rabbit anti-SOX9 ( Millipore , # AB5535 , diluted 1:200 ) , mouse anti-acetylated tubulin ( Sigma #T6793 , diluted 1:500 ) , mouse anti-GM130 ( BD Biosciences #610822 , diluted 1:200 ) , rabbit anti-KI67 ( Novocastra NCL-Ki67p , diluted 1:250 ) , rabbit anti-γH2AX ( Ser139 ) ( 20E3 ) ( Cell Signaling Technology #9718 , diluted 1:500 ) , and rabbit anti-c-KIT antibody ( Cell Signaling Technology , # 3074S , diluted 1:400 ) . Sections were washed in PBS , and incubated with the following secondary antibodies at room temperature: goat anti-rabbit biotin-conjugated secondary ( Jackson ImmunoResearch Laboratories Inc . , #111-067-003 , POC1A , PLZF , SOX9 , Ki67 , γH2AX , and c-KIT for 1 hour ) , or the biotinylated anti-mouse secondary included in the M . O . M Kit from Vector Laboratories according to the manufacturer’s instructions . The subsequent steps were carried out according to the instructions provided in the TSA Fluorescein Tyramide Kit ( TSA-FITC Kit ) by Perkin-Elmer . Sections were counterstained with DAPI to reveal nuclei , cover slipped and photographed with a Leica Leitz DMRB/E compound microscope . Quantitation was done on three animals per genotype and age on 5–10 sections per individual and presented ± std . dev . Mouse embryonic fibroblasts ( MEFs ) were isolated from individual embryos at embryonic day 13 . 5 by the University of Michigan Transgenic Animal Model Core , and stocks were genotyped for Poc1acha/cha and frozen . MEFs were thawed , grown on gelatin-coated coverslips for two days , then serum treated or serum starved for 24 hours and fixed for 20 minutes in 4% paraformaldehyde at room temperature . The coverslips were washed in PBS , permeabilized in 0 . 1% SDS for 10 minutes at room temperature , and were incubated with the same primary antibodies for acetylated tubulin and GM130 listed above . The coverslips treated with acetylated tubulin primary antibody were then incubated in an Alexa 488 conjugated anti- mouse secondary ( Life Technologies , # A-21141 ) . The coverslips incubated with the GM130 primary were incubated with the M . O . M . anti-mouse Biotinylated IgG included in the M . O . M . Kit ( Vector Laboratories ) . The GM130-incubated coverslips were then treated in the same way as the tissue sections according to the TSA-FITC Kit ( Perkin-Elmer ) . All coverslips were counterstained with DAPI and photographed using an Olympus FluoView Laser Scanning Confocal Microscope ( Microscopy and Image Analysis Laboratory , University of Michigan , Ann Arbor , MI ) . Sections through P15 tibiae were de-waxed in xylene , rehydrated through an ethanol series to 1 X PBS , and then pretreated by incubating the slides in citric acid ( pH 6 . 0 ) at 65°C for 1 hour . Slides were washed in PBS and treated according to the manufacturer’s instructions thereafter ( Roche In Situ Cell Death Detection Kit , Fluorescein , #11684795910 ) . Quantification was done on three separate sections from two different mice of each genotype and presented as average per section ± standard error . The two-tailed T test was applied to assess significance . Rosa-Lac Z mice ( Jackson Laboratories , Bar Harbor , ME , #002073 ) were crossed to Poc1acha/cha homozygous females to generate reporter positive heterozygous mice for intercrossing and generation of reporter positive wild type and mutant males for the experiments requiring labeled mutant germ cells . Spermatogonial stem cell transplantation was performed as described previously [77] . Briefly , single cell suspensions from donor testes were generated by two-step enzymatic digestion and spermatogonia were enriched by selection with a 30% continuous Percoll gradient . Adult wild type recipient mice were prepared by busulfan treatment ( 60 mg/kg of body weight ) at least 6 weeks before transplantation to deplete endogenous germ cells as described previously [77] . Adult Poc1a mutant recipient mice were not prepared with busulfan treatment because the endogenous germline was already depleted . For experiments involving Poc1a mutant and wild type counterparts as SSC donors , NCr nude mice ( Taconic ) were used as recipients to avoid immunological incompatibility . For all transplantations , donor cells were re-suspended in mouse serum-free injection media [78] at 1X106 cells/ml and approximately 10 μl of cell suspension was microinjected into the seminiferous tubules of each recipient testis . The recipient testes were examined for donor-derived colonies of spermatogenesis via X-Gal staining ~3 months after transplantation . The molecular structure of mouse POC1A was predicted using the WD40 structure predictor algorithm ( WDSP ) [34] and models generated using PyMOL ( The PyMOL Molecular Graphics System , Version 1 . 7 . 2 . 2 Schrödinger , LLC ) .
|
Severe short stature in humans has many causes including defects in skeletal and hormonal growth regulation . Primordial dwarfisms are congenital growth defects that involve mutations in genes for DNA repair , DNA replication , splicing of U12 introns , and centrosome dynamics . We discovered that the spontaneous , dwarf mouse mutant chagun is caused by loss-of-function of the gene Poc1a , which encodes protein of the centriole 1A . Mutants exhibit disproportionate dwarfism , and bones formed by endochondral and intramembranous ossification processes are affected . The epiphyseal growth plates of their long bones are disorganized . The chagun males are infertile due to Sertoli cell dysfunction and the failure of germ cells to complete meiosis . The chagun mouse is a model for human dwarfism and provides insight into the mechanism whereby this centriolar protein affects bone growth and spermatogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
LINE-1 Mediated Insertion into Poc1a (Protein of Centriole 1 A) Causes Growth Insufficiency and Male Infertility in Mice
|
Entry of herpes simplex virus ( HSV ) into a target cell requires complex interactions and conformational changes by viral glycoproteins gD , gH/gL , and gB . During viral entry , gB transitions from a prefusion to a postfusion conformation , driving fusion of the viral envelope with the host cell membrane . While the structure of postfusion gB is known , the prefusion conformation of gB remains elusive . As the prefusion conformation of gB is a critical target for neutralizing antibodies , we set out to describe its structure by making genetic insertions of fluorescent proteins ( FP ) throughout the gB ectodomain . We created gB constructs with FP insertions in each of the three globular domains of gB . Among 21 FP insertion constructs , we found 8 that allowed gB to remain membrane fusion competent . Due to the size of an FP , regions in gB that tolerate FP insertion must be solvent exposed . Two FP insertion mutants were cell-surface expressed but non-functional , while FP insertions located in the crown were not surface expressed . This is the first report of placing a fluorescent protein insertion within a structural domain of a functional viral fusion protein , and our results are consistent with a model of prefusion HSV gB constructed from the prefusion VSV G crystal structure . Additionally , we found that functional FP insertions from two different structural domains could be combined to create a functional form of gB labeled with both CFP and YFP . FRET was measured with this construct , and we found that when co-expressed with gH/gL , the FRET signal from gB was significantly different from the construct containing CFP alone , as well as gB found in syncytia , indicating that this construct and others of similar design are likely to be powerful tools to monitor the conformation of gB in any model system accessible to light microscopy .
Herpes simplex virus infections are common , with severe disease striking some individuals while others are asymptomatic . Herpes simplex virus type 1 ( HSV-1 ) afflicts roughly 70% of individuals within the United States [1] , while HSV-2 afflicts approximately 16% [2] . Infection is chronic , with some individuals suffering only mild symptoms , while others experience frequent recurrence of viral lesions [3] . Ocular HSV infection can result in scarring of the eye , resulting in blindness if untreated [4] . Herpes encephalitis , most common among newborns , has a 20% mortality rate , although almost all survivors suffer neurological abnormalities [5] . Individuals with asymptomatic infection still shed significant amounts of virus , continuing to put others at risk [3] . HSV is an enveloped virus and enters a host cell by membrane fusion , either at the plasma membrane [6] , [7] , or through endocytosis [8] depending on the target cell type . Fusion requires the HSV fusion protein gB , and accessory proteins gH/gL and gD [9] , [10] . gB and gH/gL are conserved throughout all herpes viruses , and these proteins have been described as the “core fusion machinery” [11] . Receptor binding protein gD is unique to most alphaherpesviruses , yet functionally equivalent proteins are found in the other two subfamilies of herpes viruses [11] . For HSV , gD binds a cellular receptor , such as Nectin-1 or HVEM , causing conformational changes that initiate a cascade of events leading to membrane fusion [12] . gD does so by activating regulatory protein gH/gL , which triggers gB to catalyze membrane fusion [13] . gB is the only essential HSV glycoprotein that must be membrane anchored in order to function , therefore confirming it as the viral fusogen [12] , [14] . Viral fusion proteins such as HSV gB undergo large conformational changes to fuse host and viral membranes . Often regions of the structure essential for conformational change are buried in the protein fold , or obscured by glycan which can shield the protein structure from neutralizing antibodies [15] , [16] . Knowledge of these conformational changes in other viruses have revealed mechanisms for certain neutralizing antibodies [17] , [18] , and have also led to therapeutics that target and inhibit viral entry [19] , [20] . Conformational changes of viral fusion proteins consist of at least three states; a compact prefusion form , an extended intermediate form that spans both viral and host membranes , and a final postfusion form where the fusion protein has fused host and viral membranes by bringing together the fusion loops and transmembrane domain of the fusion protein [21] . Finer molecular detail has been difficult to resolve due to the topological constraints posed by viral and target membranes , as well as the heterogeneity of intermediate states . Nonetheless , an understanding of the fusion protein refolding process is tantamount to identifying vulnerable regions of fusion proteins for design of both vaccines and therapeutics . The structure of HSV gB [22] is conserved not only among other herpes viruses such as Epstein-Barr virus [23] , but also across more distantly related viruses such as baculovirus gp64 [24] and vesicular stomatitis virus protein G [25] , [26] , all of which have been termed class III fusion proteins [27] . The crystal structure of HSV gB reveals a tall trimeric molecule with a long , helical central stalk ( one protomer of which is shown in Fig . 1 ) . [22] . The structure has been divided into four functional regions ( FR ) ( Fig . 1 ) , identified by four major immunogenic regions on HSV gB that are targets for virus-neutralizing antibodies [28] . FR1 is the first globular domain in the amino acid sequence , and contains the two fusion loops which form a fusion domain that mediates lipid binding by gB [29] . FR2 contains epitopes for neutralizing antibodies that block association with gH/gL , suggesting that important regulatory interactions with gH/gL are mediated by this region [14] . Following FR2 is an extended alpha helix that , together with the other subunits of the homotrimer , forms a central coiled-coil , also known as a central stalk . Trimerization of gB is mediated largely by contacts within the central stalk , which is a stable interaction as demonstrated by the thermostability of the trimer [30] . This extended central stalk brings together elements of the structure containing the fusion loops and the transmembrane domain , and thus is a hallmark of the postfusion state [19] . The C-terminus of the ectodomain packs against the central coiled-coil of the central stalk , and is believed to drive the fold-back that mediates membrane fusion [22] . Mutations in the C-terminus of the ectodomain have been shown to be hypofusogenic [31] . FR3 sits between components of the central stalk , and has been termed the “crown” due to its appearance in the crystal structure [22] . FR3 has also been described as a hinge due to its position in the sequence . The hinge action of the crown has been proposed to allow the fusion loops to extend towards the target cell and away from the transmembrane anchor in the viral envelope [22] . A membrane proximal region follows the C-terminus of the ectodomain and connects the ectodomain to the transmembrane anchor . The membrane proximal region has been shown to block the fusion loops from interacting with membrane [32] . The N-terminal 70 amino acids of the ectodomain , known as FR4 , are not resolved in the crystal structure , likely because they are very flexible . The single example in the literature of a large insertion in the gB ectodomain is at position 42 ( in the unstructured N-terminal region ) , where GFP was inserted without interfering with gB function [33] . This fluorescent protein construct , as well as others in the cytoplasmic tail of gB , facilitated study of trafficking and packaging of gB into the viral envelope [34] . Mutagenesis studies have been used to identify functionally important regions of gB , which are found throughout the ectodomain . Random linker-insertion mutagenesis , using linkers of five amino acids , revealed functional insertion constructs within the N-terminus , and within a loop in FR2 [35] . A separate study utilizing insertions of just two alanines described more sites that tolerated the smaller insertion , including sites in the N-terminus , FR1 , and FR2 [36] . Still , the majority of the insertions in that study were functionally impaired; leading to the notion that gB was extraordinarily fragile to mutations . Previous attempts at creating insertions in gB were done before knowledge of the gB crystal structure . We proposed that we could use the known structure of gB to create functional fluorescent protein ( FP ) constructs in the gB ectodomain . Because ongoing work to capture the gB prefusion structure has thus far been unsuccessful [30] , we designed the FP insertion constructs with two goals: 1 ) to characterize structural features of the prefusion form of gB and 2 ) to create fluorescent tools for future studies . We created FP insertions at eighteen distinct sites in gB and found that eight were functional in fusion , some of which were temperature sensitive . Because of the size of the FP , we believe that these eight insertion sites are likely surface exposed in prefusion gB , and are also unlikely to be involved in critical interactions within the gB molecule or with regulatory proteins such as gH/gL . Functional insertion sites were located within two different functional regions of the gB ectodomain , as well as at multiple locations within the disordered N-terminus . To our knowledge , this is the first report of a fluorescent protein insertion within a core structural domain of any viral fusion protein . In working towards an assay to directly observe gB undergoing conformational change during fusion , we recombined two of our functional constructs to create gB with insertions of both CFP and YFP . We found this construct was functional , and produces a FRET signal when co-expressed with gH/gL . This construct and others like it provide a promising approach to understanding both the nature and timing of conformational changes in gB that drive membrane fusion .
Based upon the postfusion crystal structure of HSV gB [37] , sites were selected that were likely to accommodate fluorescent protein insertions . Residues were chosen based upon proximity to large regions of bulk solvent . Further consideration was given to the secondary structure of the selected sites by favoring loops and sequences that linked defined structural domains . We selected residues 81 and 470 based upon previous results obtained for linker insertion mutants that retained function [35] . Further mutations were selected across the surface of gB in an attempt to sample all structural regions that were exposed in the crystal structure ( Fig . 1 ) . The central helical stalk and C-terminus of the ectodomain were not targeted , because we reasoned that fluorescent protein insertions in these regions would sterically disrupt the known crystal structure . Each gB fluorescent protein insertion ( gB-FP ) was created by inserting an AvrII restriction site immediately following the residue index of the insertion's given name . The AvrII restriction site introduced a linker of 2 amino acids: Pro-Arg . Either Cerulean fluorescent protein ( CFP ) , or Venus fluorescent protein ( YFP ) , was ligated into the AvrII site , yielding a two amino acid linker of proline-arginine on either side of the fluorescent protein . For the CFP insertion construct at position 81 ( gB-FP-81CFP ) , an additional glycine-serine linker was included ( see methods ) . In total , 21 fluorescent protein insertion constructs were created , that corresponded to 18 unique sites ( Fig . 1 ) and spanned the entire surface of the crystal structure , as well as the disordered N-terminus . Before assessing the function of gB-FP constructs , we first tested them for cell surface expression by CELISA , as this requires proper folding and trafficking of gB . Plasmids for each of the fluorescent gB constructs were transfected into B78H1 cells and incubated at either 37°C or 32°C for 24 h or 48 h , respectively . Cells were then fixed , and probed with a mixture of antibodies selected to target a diverse set of epitopes throughout gB . The antibody mixture included MAb A22 which binds FR3 [28] , MAb SS55 which binds FR1 [28] , MAb SS67 which binds FR3 , and polyclonal IgG R68 which was raised against denatured gB . When the cells were incubated at 37°C , all of the FR4 insertions exhibited substantial surface expression , as did the insertion at 481 in FR2 ( Fig . 2 ) . Insertions at position 52 and 81 ( FR4 ) were expressed at approximately 80% of WT , while insertions 95 and 100 ( also FR4 ) were expressed at approximately 65% of WT . Among the insertions located elsewhere in gB , only the insertion at amino acid 481 ( FR2 ) had appreciable ( 60% of WT ) surface expression . Therefore , at 37°C , only regions FR2 and FR4 tolerated FP insertions without seriously disrupting the global fold of gB . As many of our FP constructs were not surface expressed at 37°C , we next considered if culturing cells at a lower temperature would improve surface expression . We repeated the cell surface expression assay , but incubated the cells at 32°C post-transfection and extended the growth time to account for the lower temperature . For constructs that showed some cell surface expression at 37°C , incubation at 32°C increased cell surface expression relative to WT ( Fig . 2 ) . Importantly , we also observed cell surface expression for five constructs that appeared null at 37°C ( Fig . 2 ) . Surface expression levels for a number of gB-FP constructs were indistinguishable from WT , including insertions at 51 , 81 , and 95 . FP Insertions at positions 100 , 470 , and 481 were all approximately 80% of WT . gB-FP constructs 304 and 361 were both expressed at approximately 60% of WT , and 241 was expressed at 50% of WT . These last three insertions reside in FR1 , which contains the fusion loops and is a critical domain for fusion [29] . Thus , a number of FP insertion constructs , particularly those in FR1 and FR2 , were surface expressed predominantly at lower temperature . We next used immunofluorescence assay ( IFA ) with non-permeabilized cells to visually confirm surface expression of all the gB-FP constructs . Since CELISA results indicated that all fluorescent gB constructs had improved cell surface expression compared to WT at 32°C , IFA was performed exclusively at 32°C . We used MAb A22 , which binds gB with high specificity , to visualize gB on the cell surface . IFA results indicated that WT gB was expressed on the plasma membrane , while vector transfected cells were not bound by MAb A22 ( Fig . 2B ) . In agreement with results from CELISA , gB-FP constructs 52 , 81 , 95 , and 100 from the N-terminus were observed on the cell surface; as were FR1-localized gB-FP constructs 241 , 304 , and 361; and FR2-localized constructs 470 and 481 . Additionally , we observed definitive surface expression for construct gB-FP-664 by IFA , although the number of these cells observed by IFA was small , likely explaining the low level of surface expression observed by CELISA . For FP insertions within FR3 , a remaining concern was that these insertions might disrupt the A22 epitope in FR3 [28] , leading to false conclusions about surface expression . To address this concern , we re-examined FR3 localized gB-FP constructs ( 546 , 608 and 630 ) using rabbit polyclonal antibody R68 , which was raised against full-length gB . R68 confirmed the conclusion based upon MAb A22 that these constructs were not surface expressed ( Fig . S1 ) . We tested gB-FP constructs for activity using a cell-cell fusion assay based upon a split luciferase reporter [38] . Two cell populations were prepared , one expressing the HSV receptor Nectin-1 and transfected with a plasmid containing half of Renilla luciferase , and the other transfected with plasmids encoding HSV glycoproteins gD , gH/gL , and gB along with a plasmid for the other half of Renilla luciferase , as previously described [39] . When the two cell populations were mixed , cell-cell fusion was initiated , leading to the immediate reconstitution of luciferase . Using the cell permeable Renilla luciferase substrate EnduRen , fusion in living cells was monitored in real time . For cells cultured at 37°C for 24 h post-transfection , we found cell-cell fusion activity from 80–100% of WT for each of the FR4-localized gB-FP constructs ( Fig . 3 ) . Construct gB-FP-100 , which lies closest to FR3 , had activity 30% of WT , while none of the FR3 insertions had activity . gB-FP constructs 470 and 481 had activity of 50% and 70% of WT , respectively . Therefore these insertions allowed for gB function when expressed at 37°C , without disrupting either the global folding or function of HSV gB . As many of the fluorescent protein insertions in gB had improved surface expression at 32°C relative to WT , we next asked if fusion activity for these constructs also improved at the lower temperature . We therefore repeated the split luciferase cell-cell fusion assay with cells cultured at 32°C post-transfection for 48 h , and we report the results as a percentage of WT activity at 32°C . Some constructs , but not all , had increased fusion activity at 32°C . gB-FP constructs in FR4 increased slightly in activity , approaching or slightly exceeding WT activity ( Fig . 3 ) . FR2 localized construct gB-FP-470 improved from 50% to 70% . Construct gB-FP-304 ( FR1 ) was expressed on the cell surface only at 32°C , and gained fusion activity up to 40% of WT at this temperature . Construct gB-FP-304 is located in the same domain as the fusion loops; therefore fusion activity by gB-FP-304 was of particular interest . Fusion activity of gB-FP-361 , also in FR1 , was only 4% of WT at the 6 h endpoint , despite being expressed on the cell surface . While both gB-FP constructs 241 and 664 were surface expressed at 32°C as observed by IFA , neither of these constructs had fusion activity that could be distinguished from background . Lack of fusion by gB-FP-664 may be completely due to a low level of surface expression , since surface expression measured by CELISA for that construct was minimal . On the other hand , gB-FP-241 accumulated to 50% of WT on the cell surface , and lack of fusion activity likely represents a genuine functional impairment after reaching the cell surface . Constructs gB-FP-100 and gB-FP-481 decreased in fusion activity when the assay was done at 32°C as opposed to 37°C , suggesting that the rate limiting step in fusion for these two constructs was fundamentally different from the others , as they exhibited temperature dependence that was opposite from the others . In summary , by expressing the gB-FP proteins at 32°C , we observed fusion activity for gB-FP constructs in FR1 , FR2 , and FR4 of HSV gB , of which FR1 is notable for binding target cell membrane via the fusion loops . To confirm the luciferase assay results described above , we used fluorescence microscopy to visualize syncytium formation , a variant of the classic cell-cell fusion assay [40] . Cells were prepared that expressed Nectin-1 and the full complement of HSV fusion proteins: gD , gH/gL , and gB . After a 24 h growth at 37°C , cells were fixed and prepared for light microscopy . In agreement with split luciferase results , syncytia were observed for gB-FP constructs 52 , 81 , 95 , 100 , 470 , and 481 ( data not shown ) . Since lower temperature was permissive to surface expression for several gB-FP constructs , the assay was repeated at 32°C , and combined with an increase to 96 h of protein expression . Syncytia were readily observed for all constructs that were active at 37°C , plus insertions 304 , and 361 ( Fig . 4 ) . The extended interval for protein expression is likely to have compensated for the slow rate of fusion for gB-FP constructs 304 and 361 that were observed in the split luciferase kinetic assay . At 96 h post-transfection we are observing the endpoint of cell-cell fusion at 32°C; as essentially no further activity was observed at a 120 h time point . This provides a comparison to the split luciferase assay performed 48 h post-transfection for 32°C expression , when protein was beginning to accumulate in measurable concentrations at the cell surface . Other constructs that were inactive in the split luciferase assay were confirmed to be inactive by light microscopy at this 96 h time point , including insertions 241 and 664 which were expressed on the cell surface at 32°C . Surface expression and cell-cell fusion activity for all gB-FP constructs is summarized in Table 1 . Having found functional FP insertions in three different regions of gB , we asked if multiple fluorescent protein insertions could be combined into a single construct . We selected FP insertion constructs gB-81C and gB-470Y that were individually functional at 37°C . Insertions from these constructs were combined to create gB-81C-470Y , which contained two fluorescent proteins per protomer , or six fluorescent proteins per trimer . The fusion activity of this construct was approximately 20% of WT at 37°C , consistent with its poor surface expression ( Fig . 5 ) . When we expressed the dual-labeled construct at 32°C , we found both improved surface expression and an increase in activity to 60% of WT , suggesting that a limiting factor was protein folding and transport . We propose that protein folding improved at this lower temperature , thereby permitting an increase in trafficking to the cell surface and fusion activity . Thus the insertion of six FP molecules of the trimer did not disrupt the fusogenic activity of HSV gB . Here we have demonstrated that a single gB molecule can contain both fluorescent labels commonly employed in FRET studies ( Fig . 5E ) , setting a precedent that dual-labeled fusion proteins can be generated that possess fusogenic activity . An FP insertion occludes a region on the surface of gB that is similar in size to an antibody epitope , so we asked if the FP insertions blocked antibody binding to gB . We analyzed cell lysates from transfected cells using native western , and probed constructs gB-81C , gB-470Y , and the dual-labeled construct gB-81C-470Y with various antibodies whose epitopes are known in the postfusion conformation [28] , [41] . The calculated molecular weight for monomeric gB is 97 kDa , but addition of glycans during trafficking increases the observed mass to approximately 110 kDa [42] . A fluorescent protein is 27 kDa , increasing the expected mass of gB-FP constructs to between 124 and 137 kDa . In the native western , two bands dominate for gB , a high molecular weight band which corresponds to gB trimer , and a lower molecular weight band which likely represents monomer . First we probed the gB-FP constructs with an αGFP antibody . As expected , αGFP reacted with both monomeric and trimeric forms of each gB-FP construct , but neither with WT , nor with vector-only transfected cells ( Fig . 6A ) . We next examined two antibodies with epitopes known to localize to FR2 and which were likely to be affected by the insertion at position 470 . MAb H1781 binds a linear epitope at position 454–473 [43] , and MAb C226 binds a continuous epitope that contains residue 414 [28] , [41] . Both of these MAbs bind and neutralize virus . H1781 bound to both WT gB and gB- 81C , while H1781 did not react with either gB-470Y or gB-81C-470Y , indicating that the epitope for H1781 was either obscured or altered by the fluorescent protein at position 470 . Next , we probed the three proteins with MAb C226 and found that , unlike H1781 , C226 bound to constructs gB- 470Y and gB-81C-470Y . Therefore FR2 was only partially obscured by the fluorescent protein insertion at 470 , while the epitope involved in C226 neutralization remained accessible . We used a capture ELISA to determine the extent of similarity between WT and FP insertion constructs of gB . Polyclonal antibody specific for the cytoplasmic tail of gB was used to capture gB from cell lysates on an ELISA plate , and then the captured gB was probed with MAbs DL21 ( conformational , competes with MAbs localized to the crown ) , SS55 ( trimer specific , localized to FR1 ) , DL16 ( trimer specific , localized to FR1 ) , SS10 ( FR3 specific ) , and SS144 ( FR1 specific ) [28] , [41] . Reactivity of MAbs with each of the FP insertion constructs and WT gB was similar , indicating that MAb epitopes were displayed on the FP insertion constructs equivalently to WT . This is consistent with each of the constructs adopting a similar tertiary structure , in spite of the FP insertions in 81C , 470Y , and the double insertion in 81C470Y ( Fig . 6B ) . Included in this observation are MAbs that bind epitopes distant from the FP insertion sites , such as SS55 and SS144 , as well as epitopes near the 470 insertion , such as C226 . We conclude from the antigenic mapping that fluorescent protein insertions at these two sites did not broadly disrupt gB tertiary structure . Fluorescence resonance energy transfer ( FRET ) is a powerful technique that can detect if two fluorescent proteins are in close proximity to each other [44] . To establish if FRET can be measured with fluorescent gB constructs , we applied fluorescence lifetime imaging microscopy ( FLIM ) to our dual-labeled gB construct . Four different conditions were prepared for FRET analysis . We used the single Cerulean insertion construct gB-81C as a negative control , which accounts for Cerulean being both membrane anchored and present as a homotrimer through its linkage to gB . We compared gB-81C to dual-labeled Cerulean-Venus construct gB-81C-470Y expressed alone , or in conjunction with gH/gL . In a fourth condition , we induced cell fusion by adding soluble gD protein to cells transfected with gB-81C-470Y and gH/gL [14] . Proteins were expressed through transfection of B78H1-C10 cells ( expressing Nectin-1 ) as described in surface expression and fluorescence assays , and live cells were imaged using time-correlated single photon counting , which measures the fluorescence decay at each pixel of an image . Each pixel is fit to an exponential decay , yielding the fluorescence lifetime . If the donor fluorophore , Cerulean in this case , is undergoing FRET with an acceptor fluorophore such as Venus , then the observed fluorescence lifetime of the donor fluorophore is decreased , and this commonly indicates that the distance between donor and acceptor fluorophores has decreased . Observed fluorescence decay curves describing individual cells did not fit a single exponential decay ( Fig . 7 ) , suggesting that more than one fluorescence lifetime was being observed within a single cell in all of the conditions analyzed . The distributions of fluorescence lifetimes observed for Cerulean within a cell are depicted in a fluorescence lifetime histogram ( Fig . 7C , S2 ) . It was not possible to identify compartments within the cell that showed distinct fluorescence lifetimes , so the regions of interest selected for the fluorescence decay curves and fluorescence lifetime histograms were chosen to represent the whole cell , and each cell was treated as an independent observation . Our efforts to decompose the fluorescence decay curves into the sum of two exponentials did not consistently result in two distinct lifetime values that had relevance to Cerulean , and was not pursued . A similar effort to fit the donor lifetime histogram to the sum of two Gaussian curves revealed that components with lifetimes much longer than expected for Cerulean were disrupting the curve fitting . A single Gaussian provided a reasonable fit to the peak of the donor lifetime histogram for most cells , while a few cells ( one from gB-81C , one from gB-81C-470Y+gH/gL , and two from soluble gD ) were clearly not described by this curve fit . Therefore we selected the peak of the donor lifetime histogram ( the mode of the lifetime distribution ) to represent the Cerulean fluorescence lifetime for the cell as a whole , and we found this was the best representation of the fluorescence lifetime for the most prevalent fraction of gB molecules within a cell ( Fig . S2 ) . Cells were ranked by their donor fluorescence lifetime , and cells that represented the median of the observed lifetimes for each construct were selected for Fig . 7 A–C . Fluorescence decay curves indicated modest differences , with the control having relatively slow decay ( Fig . 7B ) . gB-81C-470Y initially had faster decay than the control , but at later time points within the decay curve , the slope of gB-81C-470Y was similar to the control , suggesting a subset of these proteins had shorter lifetimes , while another subset had lifetimes similar to gB-81C . The addition of gH/gL to gB-81C-470Y yielded the fastest decay curve at early time points , but also the largest curvature of the decay curve , indicating heterogeneity of fluorescence lifetimes within the cell , and therefore heterogeneity of either the gB conformation , or the local environment of Cerulean within the cell . When cell-cell fusion was induced by adding soluble gD to cells expressing gB-81C-470Y , gH/gL , and HSV receptor Nectin-1 , the slope of the fluorescence decay curve was reduced . This suggested that even when selecting whole cells as our region of interest , we may be observing a net change in fluorescence lifetimes for the donor fluorophore Cerulean when converting gB from prefusion to postfusion . Histograms of the Cerulean fluorescence lifetime reveal the distribution of fluorescence donor lifetimes within a cell . Comparing our four different conditions measured by FLIM-FRET , the peak of the fluorescence lifetime histograms were similar , except when gB was coexpressed with gH/gL , where the histogram peak was shifted to a shorter lifetime ( Fig . 7C , S2 ) . We calculated the mean and standard deviation the fluorescence lifetime histogram for all cells from each construct , which were 2 . 37±0 . 17 ns for gB-81C , 2 . 31±0 . 17 ns for gB-81C-470Y , 2 . 15±0 . 06 ns for gB81C-470Y co-expressed with gH/gL , and 2 . 43±0 . 11 ns for gB-81C-470Y co-expressed with gH/gL and supplemented with soluble gD to induce fusion . Given the number of cells that were observed , we needed a statistical test to evaluate if these differences were significant . We used Tukey's HSD test in conjunction with an ANOVA to calculate multiplicity adjusted P-values describing the likelihood that each condition represented a different fluorescence lifetime distribution . Tukey's HSD test accounts for the multiple comparisons between each of the four conditions . We found that a P-value of 0 . 030 described the difference between the control and gB-81C-470Y expressed with gH/gL , while the difference between the control and gB-81C-470Y alone had a P-value of 0 . 14 ( Fig . S3 ) . Therefore , co-expression of gH/gL with gB-81C-470Y resulted in a statistically significant reduction in Cerulean lifetimes when compared to the gB -81C control , while gB-81C-470Y alone did not . The reduced fluorescence lifetimes indicated that positions 81 and 470 in prefusion gB are closer together when in the presence of gH/gL . This is direct evidence that gH/gL was influencing the conformation of gB , which , to our knowledge , has not been demonstrated previously . The effect of gH/gL on gB in the absence of gD was not anticipated , but this state most likely represents the prefusion conformation . Since gB-81C-470Y expressed alone was not significantly different from gB-81C , we are forced to conclude that gB-81C-470Y expressed alone must be in a “pre-prefusion” structure , as previous results expressing gB alone resulted in cell-cell fusion after gH/gL was provided in trans or as a soluble protein [12] . When we induced cell-cell fusion by adding soluble gD to cells that were co-transfected with gB-81C-470Y and gH/gL , fluorescence became more diffuse when spread across syncytia . Nonetheless , we applied the same analysis to these cells , finding that the mean lifetimes were statistically different from cells that did not have added gD , with an adjusted P-value of 0 . 0058 ( Fig . 7D ) . We interpreted the increase in fluorescence lifetime of Cerulean found in syncytia to mean that the distance between Cerulean and Venus in the gB-81C-470Y construct increased as a result of cell-cell fusion . Therefore the change in FRET was a reporter for conformational change associated with fusion . To ensure that the observed change in FRET was strictly due to interactions between Cerulean and Venus , we performed a negative control to determine if the fluorescence lifetime of gB-81C was affected by the coexpression of gH/gL , or by initiation of fusion with soluble gD . We repeated the FLIM-FRET analysis for these two conditions in addition to gB-81C alone , but we observed no change in fluorescence lifetime compared to gB-81C ( Fig . S4 , Fig . S5 ) . While these new conditions were collected and analyzed independently , we found it informative to consider all the data at once , juxtaposing cells expressing gB-81C with those expressing gB-81C-470Y ( Fig . S4E ) . If the same statistical analysis were performed across all of the FLIM-FRET data , we note that the two comparisons found to be statistically significant in our initial studies would remain so . An increase in the total number of statistical comparison without an increase in amount of data resulted in a slight increase in the P-value for comparisons between gB-81C-470Y coexpressed with gH/gL and and gB-81C-470Y with gH/gL and added soluble gD . The adjusted P-value decreased for the difference between gB-81C and gB-81C-470Y coexpressed with gH/gL , due to the additional data for gB-81C affirming the differences ( Fig . S6 ) .
Viral fusion proteins such as gB undergo large rearrangements of tertiary structure caused by conformational changes during viral fusion and entry . These structural changes are largely uncharacterized for HSV gB , as only the postfusion crystal structure is known . We designed fluorescent protein insertions in HSV gB to illustrate which gB domains will tolerate the steric bulk of an insertion of this size . If the insertion is functional , the result shows that the FP occupies a solvent exposed region on the surface of gB , i . e . , one that is not buried by the tertiary arrangements of structural domains . Furthermore , the function of such a construct shows that this region is exposed at all steps of fusion , thus giving us information about structures for which there has yet to be a solution . Functional gB-FP constructs could also become tools for future characterization of gB structure through fluorescence-based methods . We engineered fluorescent gB constructs by selecting regions from each domain of the HSV gB crystal structure that were conspicuously surface exposed in the postfusion structure . We anticipated that many of these gB-FP constructs would be non-functional , since the prefusion structure is likely to be very different from the postfusion structure . Furthermore , essential functions have been assigned to each of the structural domains of gB where we designed fluorescent protein insertions [14] , [29] , [43] , [45] , decreasing the likelihood of creating functional constructs . Indeed we found that many of these insertions disrupted proper surface expression of gB when expressed at 37°C . However , construct gB-FP-481 , as well as several in FR4 , were successfully surface expressed ( Fig . 2 ) , and were functional at 37°C ( Fig . 3 , 4 ) . Furthermore , antigenic analysis of insertion constructs 81 and 470 indicated that the tertiary structure of these constructs or the dual-labeled construct was indistinguishable from WT ( Fig . 6B ) . This result is an important extension of previous findings that these two regions tolerate short linker insertions [35] , [36] , or domain insertions at specific sites in either FR4 or FR2 [33] , [46] . Even with small amounts of gB-470 on the cell surface ( Fig . 2B ) , activity of this construct was still 50% of WT as measured by the luciferase cell-fusion assay ( Fig . 3 ) , indicating that fusion activity for this construct was greater than anticipated by surface expression levels . By expressing fluorescent gB constructs at a more permissive temperature of 32°C , we found that surface expression improved for gB-FP-470 in FR2 , as well as for gB-FP-304 and gB-FP-361 in FR1 . Large insertions in FR1 that retain function are unprecedented . FR1 is functionally critical to gB for at least two reasons: it includes epitopes that are targeted by neutralizing antibodies [28] , and it contains the fusion loops which bind the target cell membrane [29] . FR1 is also sterically crowded , as postfusion gB molecules closely associate laterally on the membrane [47] . For these reasons , fusion activity by two fluorescent protein insertions located in FR1 should make them an important tool to dissect the mechanism of fusion by gB . Insertion constructs located in FR4 , the N-terminus of gB , were found to be surface expressed and functional , but not all N-terminal insertions had equivalent activity . gB-FP constructs 52 , 81 , and 95 all had robust fusion activity , although insertion 95 had less activity than the other two . Construct gB-FP-95 is separated by eight amino acids from FR3 , and gB-FP-100 is just three amino acids away from FR3 in the crystal structure [37] . The fusion activity of gB-FP-100 was only 10% of WT rate in the 6 h fusion assay , yet fusion activity was readily observed at later time points ( Fig . 4 ) . Construct gB-FP-100 was one of only two constructs that had higher fusion activity at 37°C relative to WT than at 32°C relative to WT . For construct gB-FP-100 in particular , it is interesting to speculate that the added steric bulk of the fluorescent protein near FR3 , which may serve as a hinge during fusion , could be impeding the transition between prefusion and postfusion conformations , and that higher temperature can overcome the steric encumbrance of the FP insertion . In total , constructs from eight distinct insertion sites in gB were demonstrated to be functional when expressed at permissive temperature of 32°C , while constructs for eight distinct insertion sites were not expressed , and two sites were surface expressed , but non-functional ( Fig . 1 , Table 1 ) . Functional gB-FP constructs inherently describe sites in gB that are not directly involved in critical interactions either within the gB molecule , across gB molecules , or with regulatory protein gH/gL . Significant constraints are placed upon a model of prefusion gB by considering both the connectivity of protein domains along with knowledge of which surfaces on these domains are solvent exposed , as determined by functional fluorescent protein insertions . Functional gB-FP constructs 304 , 361 , 470 , and 481 broadly represent the outside surface of FR1 and FR2 in the postfusion structure , and our results indicate that these sites also make up the outside surface of these regions in the prefusion structure . We have previously constructed a working model of the HSV gB prefusion structure based upon homology to vesicular stomatitis virus protein G ( VSV G ) , as well as fusion rate measurements of HSV gB point mutants [39] . Here we consider the working model of the prefusion structure again in an effort to rationalize both functional and non-functional fluorescent protein insertions . The model ( Fig . 8A , Supporting Data S1 ) was constructed with UCSF Chimera [48] by aligning domains from the HSV gB postfusion crystal structure [37] onto the prefusion crystal structure of VSV G [26] . The loops that link domains were not adjusted relative to the postfusion structure , therefore these loops often clash with neighboring domains . We argue that adjusting these loops without an appropriate template would merely obscure the shortcomings of the model while likely moving the model further from the true structure . The model was built on the assumption that tertiary structure rearrangements are the primary difference between prefusion and postfusion structures . By assessing the extent that experimental evidence supports this assumption for each of the domains , we would gain insight about the prefusion structure . The locations of functional FP insertions predicted by the prefusion model are consistent with the FP insertions facing outward towards bulk solvent in the model ( Fig . 8A ) . The model is further supported by the prediction that the glycans , located in FR2 , cluster at the top of the prefusion model , pointed away from the fusion loops and presumably outward from the viral envelope ( Fig . 8B ) . Thus , they are predicted to be positioned on the outside of prefusion gB , shielding FR2 from neutralizing antibodies . Non-functional FP insertion constructs that are surface expressed are an interesting case , because they suffer a defect in either regulation or execution of fusion . Examples of surface-expressed non-functional gB mutants have been rare , although two different insertions of dipeptides in FR2 were previously identified as such [36] . gB-FP constructs 241 and 664 ( Fig . 1 , purple spheres ) were surface expressed at 32°C ( Fig . 2 ) , but did not function ( Fig . 3 , 4 ) . Thus , these constructs could potentially either be trapped in the prefusion conformation , or emerge at the cell surface in the postfusion conformation which is incapable of fusion , and here we consider these two constructs separately . Based upon our prefusion model of gB , amino acid 241 maps to a surface exposed face of FR1 adjacent to FR3 ( Fig . 8 ) . Positioned at the side of the prefusion structure , gB-FP-241 may block lateral interactions that are known to occur between gB trimers [47] . Alternatively , gB-FP-241 may block the approach of the regulatory protein gH/gL , which binds gB as an essential step for triggering fusion [12] , [13] . As yet another possibility , construct gB-FP-241 could directly interfere with conformational changes between prefusion and postfusion structures . While we cannot conclude which mechanism causes gB-FP-241 to be fusion deficient , gB-FP-241 contains a fluorescent probe localized to a domain of gB critical to fusion regulation and activity , and may represent a trapped prefusion structure . Insertion 664 resides in a linker between the crown and the C-terminal component of the central stalk . Very low levels of surface expression of gB-FP-664 were observed by CELISA ( Fig . 2A ) ; therefore lack of fusion activity by this construct may be due to inadequate amounts of gB-FP-664 that have accumulated on the cell surface . Still , we note that a minority of cells were observed by IFA to express gB-FP-664 in appreciable amounts ( Fig . 2B ) , therefore we give additional consideration to the effect of insertion 664 on the structure . In the prefusion model , this site maps to a buried region of the crown , although the VSV G prefusion structure does not provide a template for the C-terminal component of the central stalk , and the direction of the amino acid chain following the crown , including position 664 , is poorly predicted . The closest protein surface to insertion 664 is also shared by insertion 241 of a symmetry-related protomer , raising the possibility that these two insertions affect the same interface of gB . In the postfusion structure , site 664 is tucked under the crown near the central stalk , providing less room for the FP compared to the other insertion sites that were selected . While the crown of gB ( FR3 ) seems an attractive location for insertion mutations , being solvent exposed and well separated from other domains in the crystal structure , we found no functional gB-FP constructs that localized to the crown . We had previously suggested that FR3 is buried in the center of the prefusion structure [39] . If this is so , then it is not surprising that FP insertions in FR3 would cause steric clashes with FR1 and FR2 . FR3 has also been previously reported to mediate binding to cell surfaces , in an apparent contradiction to FR3 being buried in the prefusion conformation [45] . This observation can be rationalized by noting that cell surface binding was measured with recombinantly expressed soluble gB , which is in the postfusion conformation . Therefore it has not been demonstrated that FR3 can mediate cell surface binding in the prefusion conformation . An alternative explanation is that the crown may interact with the cell surface at an intermediate step of fusion when gB is extended , after the fusion loops have engaged a target membrane , but when significant distance still separates viral and target membranes . Therefore it remains plausible that the crown forms the center of the prefusion structure , partially buried at its core . Our experimental results support this conclusion , as none of the FR3 insertions were surface expressed ( Fig . 2 ) , suggesting to us that they were unfolded and retained within the ER [49] . Fluorescence resonance energy transfer ( FRET ) measurements can provide a measure of distance between two fluorescent proteins , which is an attractive application of these gB-FP constructs . Dual-labeled , functional gB-FP constructs could be used to directly measure distance between gB domains either at the prefusion state , or in stable intermediates . Two major impediments have prevented FRET measurements in viral fusion proteins thus far . Fluorescent protein insertions have been viewed as too disruptive to viral fusion proteins that are already delicately balanced between prefusion , postfusion , and complete misfolding . With HSV gB , we have established several functional fluorescent gB constructs , and our success rate suggests that other interesting insertion sites remain to be found . Furthermore , we have created a dual-labeled fluorescent protein construct , indicating that individual insertions could be easily paired in a single construct . A second impediment to FRET studies of viral fusion proteins is caused by pH change , which is the trigger for conformation change in a number of well characterized viral fusion proteins , such as influenza hemagglutinin or VSV G [19] . Derivatives of either GFP or YFP are typically required for FRET measurements , and since GFP and YFP are quenched by low pH [50] , quantitation measurement of FRET across a varying pH is intractable . HSV gB is capable of fusing at the cell surface [9] , [51] , and therefore circumvents the major problem of pH instability during endocytosis . Taken together , HSV gB-FP constructs present a surprisingly suitable model system to study viral fusion with real-time FRET observations . We have performed preliminary FRET analysis using our gB-81C-470Y construct that contains both Cerulean and Venus . Based upon our prefusion model , we had predicted that the distance between insertion sites 81 and 470 would be shorter in the prefusion state than in the postfusion , but even if our prediction was correct , it was unclear that a difference could be observed by FRET . Using FLIM-FRET methodology , we found that there was a significant difference between our control construct gB-81C and gB-81C-470Y when the latter was co-expressed with gH/gL , indicating that FRET was occurring and the distance between sites 81 and 470 was reduced in this state ( Fig . 7 ) . We estimate an upper bound for this distance to be approximately 85 Å , based upon the Förster distance for Venus and Cerulean of 52 Å [52] , the dependence of the FRET efficiency on the inverse sixth power of the distance between fluorophores [44] , and assuming at least a 5% FRET efficiency was required for observation . Observed fluorescence lifetime values were similar between the gB-81C control and gB-81C-470Y when expressed alone , suggesting that for the majority of gB protein , FRET was not occurring when gH/gL was not present . Furthermore , the reduced Cerulean fluorescence lifetime was due to a FRET interaction with Venus , since the gB-81C construct expressed with gH/gL had similar fluorescence lifetimes to gB-81C alone ( Fig . S4 ) . When we added soluble gD to cells that were transfected with both gB-81C-470Y and gH/gL , cell-cell fusion commenced , and we observed longer fluorescence lifetimes again for Cerulean , similar to gB-81C when expressed alone . We interpret this to mean that gB transitioned from the prefusion to the postfusion state , which resulted in a loss of FRET signal . Thus , we infer that sites 81 and 470 are closer together in the prefusion state than in the postfusion ( Fig . S7 ) . A surprising feature of the FLIM measurements was the overall short Cerulean fluorescence lifetimes , even in the control construct . The fluorescence lifetime of Cerulean alone has been measured as 3 ns , while Cerulean linked directly to Venus with a 5 amino acid linker has a lifetime of 1 . 6 ns [53] . The longest observed lifetime for our Cerulean alone construct , gB-81C , was only 2 . 6 ns , and the majority of data points had shorter lifetimes . Cerulean has previously been noted to undergo ‘energy migration’ , or homo-FRET between proximal Cerulean molecules [54] , which decreases the observed fluorescence lifetime . Two explanations are possible if this phenomenon occurs in gB-81C . If the N-terminus of gB closely associates with itself in the gB trimer , then three Cerulean molecules will be in close proximity , perhaps causing homo-FRET . A more mundane explanation may be that high level expression of a membrane protein was sufficient to saturate the local environment and drive homo-FRET . While we have examined the relative fluorescence lifetimes between our constructs , we note that to obtain definitive absolute measurements of fluorescence lifetimes , or to calculate FRET efficiencies that may be related to absolute distances between sites , it is critical that FLIM measurements be performed relative to a fluorescence lifetime standard , such as fluorescein [53] , which we have not measured and is left for future work . Our results illustrate how the FRET methodology can be used to assess the conformation of the viral fusogen gB in the context of living cells . The analysis could readily be extended in future work to make time dependent observations of gB conformation , perhaps synchronized by the addition of soluble gD as demonstrated here , or even soluble gH/gL [13] . Alternatively , viral particles could be directly studied , as demonstrated previously with fluorescently labeled capsid [55] , although it remains to be demonstrated if gB-81C-470Y can be packaged in virus . In summary , we have created fluorescent protein insertions in the HSV fusion protein gB , and interpreted the steric effects of these insertions , supporting a model of the prefusion state based upon VSV G . We found that gB functional regions 1 and 2 accommodate fluorescent protein insertions and therefore maintain a similar orientation to bulk solvent between prefusion and postfusion states . We have created a homology model of prefusion HSV gB , and found this model to be consistent with experimental observations of fluorescent gB constructs . We demonstrated that two functional fluorescent proteins could be combined within a single gB construct to create a dual-fluorescent gB molecule , while retaining function . We tested for FRET in our dual-labeled construct by observing Cerulean fluorescence lifetimes , and we obtained evidence of FRET when gB was coexpressed with gH/gL , while fluorescence lifetimes of gB in the absence of gH/gL was similar to the control . We added soluble gD to induce cell-cell fusion in cells expressing the dual-labeled gB construct , and found the FRET signal was again similar to the control , indicating that FRET was no longer occurring , and suggesting to us that the FRET signal was specific for a prefusion state . Finally , we demonstrated that the decreased Cerulean fluorescence lifetime in the presence of coexpressed gH/gL was strictly due to a FRET interaction with Venus by measuring FLIM-FRET for gB-81C coexpressed with gH/gL , finding that the fluorescence lifetime of gB-81C coexpression of gH/gL was not different from gB-81C alone . Using fluorescent protein insertions , we have leveraged experimental results to make concrete observations about the structure of prefusion gB , while creating novel tools to facilitate FRET observations of viral fusion protein conformational change during fusion .
Mutations in gB were constructed from the HSV-1 KOS gB expression plasmid pPEP98 [56] , since a crystal structure is available for this sequence [22] . Site-directed mutagenesis was performed with forward and reverse mutagenesis primers ( Table S1 ) , employing some protocol modifications to circumvent PCR problems caused by the GC-rich sequence of gB . Each 25 µl mutagenesis reaction was prepared with the following components: 0 . 5 µl Phusion DNA polymerase ( New England Biolabs ) , 5 µl 5X GC Buffer , 0 . 5 µl 10 mM deoxyribonucleotides , 2 µl dimethyl sulfoxide ( DMSO ) , 0 . 5 µl 10 µg/ml template plasmid , and 15 µl H2O . The reaction mix was split into 2 parts , and 0 . 5 µl of 10 µM primer ( Forward or Reverse ) was added to each . The PCR cycle began with 3 min at 96°C , followed by three repetitions of cycling between 96°C for 30 s and the annealing temperature for 15 s . The annealing temperature was selected using the nearest neighbor annealing temperature calculated by OligoCalc [57] . During this step , primers are extended to overcome primer-primer annealing [58] . Afterward , PCR was paused and forward and reverse primer reactions were mixed before commencing the standard site-directed mutagenesis procedure of denaturation , annealing , and extension . Product was DpnI digested and ethanol-precipitated before re-suspending in H20 and transformed into T10F' E . coli ( Life Technologies ) for DNA purification and sequence validation . Cloning was performed in the pUC19 cloning vector [59] modified to accommodate XhoI and BglII restrictions sites , which were used to shuttle validated sequences into the pCAGGS expression vector [60] . To facilitate potential recombination of gB constructs , a variant of pPEP98 , named pJG1061 , was created by site-directed mutagenesis containing silent mutations in gB introducing restrictions sites KpnI at amino acid 266 , HindIII at amino acid 544 , and AflII at amino acid 812 ( Table S1 ) . The function of this construct was identical to the parent construct ( data not shown ) . FP insertions were introduced by ligation into gB constructs that had been mutagenized to provide an AvrII restriction site , using the protocol described above . Plasmids were cleaved with AvrII , digested with shrimp alkaline phosphatase ( New England Biolabs ) , and then ligated with fluorescent protein DNA fragments flanked by AvrII sites at both ends . Fluorescent protein DNA fragments with sticky ends were generated by PCR amplification and AvrII digestion from fragments of reference plasmid C32V , which contains Venus and Cerulean fluorescent proteins [53] . Fluorescent proteins were initially amplified with addition of an eight residue glycine-serine linker , which included the creation of CFP insertion construct ( gB-FP-81CFP ) . Since all other fluorescent protein insertions made in parallel with gB-FP-81CFP containing glycine-serine linkers were non-functional , the glycine-serine linker was then omitted and all other gB-FP constructs were re-created without the linker , resulting in the constructs described in this work . A complete list of primers used for each construct is shown in Table S1 . Individual clones were initially screened by PCR for correct orientation of the FP insert , followed by sequence confirmation of the construct . FP constructs are referred to by the number of the amino acid in of HSV-1 gB ( KOS ) that precedes the insertion ( plasmid names given in parenthesis ) : 52-CFP ( pJG1032 ) , 81-CFP ( pJG1024 ) , 81-YFP ( pJG1040 ) , 95-CFP ( pJG1048 ) , 100-CFP ( pJG1049 ) , 137-YFP ( pJG1050 ) , 241-YFP ( pJG1051 ) , 304-YFP ( pJG1052 ) , 334-YFP ( pJG1053 ) , 361-YFP ( pJG1054 ) , 419-YFP ( pJG1033 ) , 430-YFP ( pJG1034 ) , 458-YFP ( pJG1035 ) , 470-CFP ( pJG1038 ) , 470-YFP ( pJG1025 ) , 481-YFP ( pJG1055 ) , 546-YFP ( pJG1036 ) , 608-YFP ( pJG1056 ) , 630-YFP ( pJG1037 ) , 664-YFP ( pJG1057 ) . A double FP insertion construct was created by utilizing the gB construct containing restriction sites introduced through silent mutations ( pJG1061 , see above ) . FP insertions from gB-81-CFP and gB-470-YFP were spliced together to create construct 81-CFP-470-YFP ( pJG1026 ) . FP insertions were visualized on the gB structure using the more recent PDB 3NW8 instead of the original gB structure PDB 2GUM because more residues of gB were resolved , including the sites of insertions 334 and 470 and the orientation of the N-terminus exiting the structure . Outside of the fusion loops , the domains in all of these structures have very similar conformations . Cell-based ELISA ( CELISA ) was used to assess cell surface expression of gB as previously described [39] . Briefly , 2 . 5×104 B78H1 cells [61] per well were plated on a 96 well plate and cultured overnight . The following day , cells were transfected using fugene6 according to the manufacturer's protocol ( Promega ) . For each well , 0 . 6 ul fugene6 was combined with 50 ng empty vector DNA and 50 ng gB expression plasmid . Cells were incubated with transfection mix for 5 h at 37°C . Transfection mix was then removed , and fresh medium was added . For 37°C expression , cells were fixed after 24 h at 37°C . For 32°C expression , cells were fixed after 48 h . After formaldehyde fixation , cells were blocked with 5% non-fat dry milk in PBS with 0 . 2% tween 20 , and then incubated for 1 h with 1 µg/ml primary antibody , which was a mix of polyclonal antibody R68 [45] , and monoclonal antibodies A22 , SS55 and SS67 [28] , selected to capture gB in any antigenic state . Cells were rinsed three times with PBS , and then incubated for 1 h with a mixture of anti-rabbit and anti-mouse HRP conjugated secondary antibodies . Again , cells were rinsed three times with PBS , then 2 , 2′-azino-bis ( 3-ethylbenzothiazoline-6-sulphonic acid ) ( ABTS ) substrate was added , and the optical density at 405 nm was recorded . Cell surface expression was visualized by immunofluorescence assay ( IFA ) , as previously described [14] . 8×104 B78H1 cells were plated on glass coverslips in a 24-well plate format and grown overnight . The following day , cells were transfected with fugene6 using 2 . 4 ul fugene6 and 400 ng total DNA per well ( 100 ng gB and 300 ng pUC19 vector carrier DNA ) . The transfection mix was removed after 5 h , fresh medium was added , and then cells were grown for 96 h at 32°C followed by formaldehyde fixation . Fixed cells were incubated 30 min in 50 mM NH4Cl , and then blocked 1 h in 10% goat serum . Cells were then incubated with 10 µg/ml A22 primary antibody in 10% goat serum for 3 h at room temperature ( RT ) , rinsed three times , and then incubated with anti-mouse secondary antibody conjugated with Alexa-594 for 2 h at RT . Cells were rinsed three times with PBS and once with water , and then mounted with ProLong Gold ( Invitrogen ) anti-fade mounting medium , and photographed at 40× magnification using a C4742-95-12NRB camera ( Hamamatsu ) . Split Renilla luciferase constructs RLuc81–7 and RLuc88–11 [38] were used to quantitate fusion activity of fluorescent gB constructs . As described previously [39] , two cell populations were prepared by transfection of B78H1 cells [61] . One cell population expressed HSV receptor Nectin-1 , and RLuc88–11 . A second cell population expressed HSV glycoproteins gB , gD , gH , and gL from plasmids pPEP98 , pPEP99 , pPEP100 , and pPEP101 , respectively [56] , as well as RLuc81–7 . For expression at 37°C , cells were assayed for 24 h post-transfection before activity was measured , and for expression at 32°C , cells were assayed after 48 h post-transfection . Fusion of cells expressing HSV receptor and glycoproteins resulted in reconstitution of the split Renilla luciferase protein , which was quantitated in live cells with membrane permeable luciferase substrate EnduRen ( Promega ) . Direct observation of fusion activity was assessed using a cell-cell fusion assay to observe formation of syncytia using fluorescence microscopy . B78H1-C10 cells ( expressing HSV receptor Nectin-1 ) [61] were plated on glass coverslips and transfected as described for IFA , except that each well of the 24 well plate was transfected with 100 ng DNA of each of the four HSV glycoproteins . For 37°C assays , cells were incubated overnight in transfection mix and fixed after 24 h . For 32°C assays , the transfection mix was removed after 5 h , fresh medium was added , and then cells were grown for 96 h at 32°C before formaldehyde fixation . Cover slips mounted and imaged as described above . 293T cells were transfected using Gene Porter ( Genlantis ) with plasmid constructs expressing WT or mutant forms of gB . 48 h after transfection , cells were lysed for 1 h at 4°C in 50 mM Tris pH 7 . 5 , 150 mM NaCl , 1% NP-40 , and 1× complete protease inhibitor cocktail ( Roche ) . Lysates were clarified by centrifugation for 30 min at 20 , 000× g and subjected to native SDS-PAGE and western blotting as described previously [62] . To do capture ELISA , extracts of transfected cells were prepared as described above for native western blot analysis . Individual wells of a 96-well plate were first coated with 50 µl of 10 µg/ml polyclonal antibody R242 that was directed against the cytoplasmic tail of gB , for 2 h . Plates were then washed 3 times with PBS supplemented with 0 . 1% Tween-20 ( PBS-T ) and incubated in blocking buffer consisting of PBS-T supplemented with 5% non-fat dry milk . After 30 min , blocking buffer was removed and cell lysates , diluted 1∶10 in blocking buffer , were added for 1 h . Plates were then washed 3 times with PBS-T and incubated with 10 µg/ml MAb IgG in blocking buffer for 1 h . Plates were again washed 3 times with PBS-T and incubated for 30 min with goat anti-mouse horseradish peroxidase ( HRP ) diluted in blocking buffer . Finally , plates were washed 3 times with PBS-T , then ABTS substrate was added and optical density 405 nm was recorded . For each antibody , the signal generated from a control well containing no lysate was subtracted from the final reading . Fluorescence lifetime imaging microscopy ( FLIM ) was applied to measure fluorescence resonance energy transfer ( FRET ) on fluorescent gB constructs . B78H1-C10 cells were plated on 35 mm dishes with glass cover slip bottoms ( MatTek ) . Cells were transfected as described above , followed by a 3 day growth interval at 32°C . Soluble gD 306t [63] was added to indicated cells 4 h before imaging . FLIM measurements were made on live cells using a Leica TCS SP5 spectral imaging confocal/multiphoton microscope . Images were acquired using Cerulean excitation wavelength of 405 nm and the fluorescence emission was monitored at 452–491 nm , using a 60× objective with a 2 . 5× zoom . Time domain FLIM decay curves for Cerulean were recorded and fluorescence donor lifetime histograms were generated using SymPhoTime software ( PicoQuant ) . Further analysis was performed using statistical software R ( R Development Core Team ) . P-values adjusted for multiple comparisons were determined using ANOVA and Tukey's HSD test .
|
Viral fusion proteins undergo complicated conformational changes in order to fuse viral and host membranes during viral entry . Conformational changes between prefusion and postfusion states also allow the virus to hide critical regions of the fusion machinery from the immune system . The structure of herpes simplex virus fusion protein gB is known only in its postfusion state , while the prefusion structure is unknown . To study the prefusion state , we created fluorescent protein ( FP ) insertions within gB and tested them for fusion activity . Due to the size of the fluorescent protein insertion , regions in gB that tolerate this insertion must be solvent exposed , thereby describing structural features of the prefusion structure . We created functional gB constructs with FP insertions in two of the three globular domains of gB , while non-functional insertions in the third domain suggested that it may be buried in the prefusion structure . Additionally , we created a dual-labeled FP gB construct which we found to report on the conformation of gB before and after fusion . Using this dual-labeled gB construct , we have demonstrated how fluorescence-based methods can be used to directly study dynamics of viral fusion proteins in living cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"viral",
"envelope",
"viral",
"entry",
"viral",
"transmission",
"and",
"infection",
"virology",
"biology",
"and",
"life",
"sciences",
"microbiology",
"viral",
"structure"
] |
2014
|
Functional Fluorescent Protein Insertions in Herpes Simplex Virus gB Report on gB Conformation before and after Execution of Membrane Fusion
|
Adult T-cell Leukemia ( ATL ) is a lymphoproliferative disease of CD4+ T-cells infected with Human T-cell Leukemia Virus type I ( HTLV-1 ) . With the exception of allogeneic hematopoietic stem cell transplantation , there are no effective treatments to cure ATL , and ATL cells often acquire resistance to conventional chemotherapeutic agents . Accumulating evidence shows that development and maintenance of ATL requires key contributions from the viral protein , HTLV-1 basic leucine zipper factor ( HBZ ) . In this study we found that HBZ activates expression of Heme Oxygenase 1 ( HMOX-1 ) , a component of the oxidative stress response that functions to detoxify free heme . Transcription of HMOX1 and other antioxidant genes is regulated by the small Mafs . These cellular basic leucine zipper ( bZIP ) factors control transcription by forming homo- or heterodimers among themselves or with other cellular bZIP factors that then bind Maf responsive elements ( MAREs ) in promoters or enhancers of antioxidant genes . Our data support a model in which HBZ activates HMOX1 transcription by forming heterodimers with the small Mafs that bind MAREs located in an upstream enhancer region . Consistent with this model , we found that HMOX-1 is upregulated in HTLV-1-transformed T-cell lines and confers these cells with resistance to heme-induced cytotoxicity . In this context , HBZ-mediated activation of HMOX-1 expression may contribute to resistance of ATL cells to certain chemotherapeutic agents . We also provide evidence that HBZ counteracts oxidative stress caused by two other HTLV-1-encoded proteins , Tax and p13 . Tax induces oxidative stress as a byproduct of driving mitotic expansion of infected cells , and p13 is believed to induce oxidative stress to eliminate infected cells that have become transformed . Therefore , in this context , HBZ-mediated activation of HMOX-1 expression may facilitate transformation . Overall , this study characterizes a novel function of HBZ that may support the development and maintenance of ATL .
The accumulation of reactive oxygen ( ROS ) and nitrogen species ( RNS ) is known to induce damage to cellular structures , including genetic material . Oxidative DNA damage can result in cell cycle arrest , the induction of replicative senescence , and initiation of apoptosis [1] . To avoid these outcomes , expression of free radical- and metal-scavenging enzymes is induced in response to oxidative stress as a means of limiting cellular damage . The induction of the antioxidant response is largely regulated by the Cap’n’Collar ( CNC ) transcription factor , NF-E2-related factor 2 ( Nrf2 ) , and the small musculoaponeurotic fibrosarcoma ( Maf ) proteins , MafF , MafG , and MafK . The small Mafs are expressed from different genes; however , they display a high degree of similarity and appear to be functionally redundant [2] . Upon sensing oxidative stress , Nrf2 and small Mafs form heterodimers that activate transcription of antioxidant genes by binding various types of antioxidant response elements ( AREs ) located in the promoters and/or enhancers of these genes [2 , 3] . Though several variations of the ARE have been characterized , all contain a consensus core sequence 5’-TGA[C/G]NNNGC-3’ [2 , 3] . Notably , chromatin immunoprecipitation-deep sequencing ( ChIP-Seq ) analyses of Nrf2 and MafG genomic binding sites that Nrf2/small Maf heterodimers frequently occupy sites containing the sequence 5’-TGCTGA[C/G]TCAGCA-3’ , termed Maf recognition elements ( MAREs ) [4–6] . Though the upregulation of antioxidant enzymes is an important cancer-prevention mechanism in healthy cells , constitutive overexpression of these proteins has been reported in a variety of malignancies . Over-activation of the oxidative stress response in these situations is associated with the development of drug and radiation resistance , increased metastasis , and with poor patient outcomes [7–9] . Heme Oxygenase I ( HMOX-1 ) is one of the antioxidant proteins implicated in mediating these effects . HMOX-1 is a heme-metabolizing enzyme that is a vital component of the iron recycling system [10] . However , HMOX-1 overexpression has been observed in a variety of malignancies , and in these settings , it has been found to promote cancer cell survival and proliferation , and the onset of multi-drug resistance [11–13] . Human T-cell leukemia virus type I ( HTLV-1 ) is a human retrovirus that predominantly infects CD4+ T-cells in vivo . While most HTLV-1-infected individuals are asymptomatic carriers of the virus , HTLV-1 can cause inflammatory and lymphoproliferative diseases . Tropical spastic paraparesis/HTLV-1-associated myelopathy/ ( TSP/HAM ) is an insidious neuroinflammatory condition resulting in demyelination of the spinal cord [14] . Though HAM/TSP pathogenesis is not completely understood , evidence suggests that anti-HTLV-1 cytotoxic T-cells ( CTLs ) target infected lymphocytes that have crossed the blood-brain barrier , resulting in increased pro-inflammatory cytokine production and damage to the surrounding nervous tissue [15] . Adult T-cell leukemia ( ATL ) is a highly aggressive lymphoproliferative disorder for which there are no effective treatments [16 , 17] . Like TSP/HAM , the mechanisms that drive ATL progression are still being elucidated; however , pathogenesis is closely associated with the activities of two , pro-oncogenic viral proteins , Tax and the HTLV-1 basic leucine zipper factor ( HBZ ) [18] . Tax is known to have roles in regulating proviral gene expression [19] , promoting cellular proliferation and replication , and inhibiting apoptosis [18 , 20] . Paradoxically , it has also been demonstrated to promote the accumulation of ROS/RNS through the hyperactivation of canonical and non-canonical NF-κB pathways , the upregulation of inducible nitric oxide synthase ( iNOS ) , and through its interaction with ubiquitin-specific protease 10 ( USP10 ) [21–26] . Incidentally , the viral accessory protein p13 has also been linked to increased production of ROS , possibly through its induction of mitochondrial membrane depolarization [22 , 27] . Detrimental effects of these two viral proteins is frequently offset by inactivation of the promoter located in the 5’ long-terminal repeat ( LTR ) of the provirus , which silences expression of all sense strand-encoded proviral genes [28] . However , recent evidence suggests that stressful conditions promote brief reactivation of transcription from the 5’ LTR promoter [29–32] . Therefore , limiting the damage caused by ROS appears to be a lifelong concern of an infected T-cell . In contrast to all other HTLV-1 proteins , HBZ is expressed from a gene encoded on the antisense strand of the provirus , leading to its constitutive expression throughout HTLV-1 disease progression [33–36] . Certain pro-survival functions of HBZ have been characterized , and its expression is critical for maintaining proliferation of HTLV-1-infected cells [35 , 37 , 38] . Notably , HBZ has been reported to prevent the induction of apoptosis [39 , 40] , rescue host cells from NF-κB-induced senescence [41 , 42] , and promote evasion of Tax-specific CTLs possibly by downregulating sense proviral transcription [43] . These reports confirm that HBZ plays a variety of cytoprotective roles within the host cell , possibly as a means of promoting long-term persistence . However , its involvement in response to Tax and p13-induced oxidative stress is unknown . We questioned whether HBZ modulates the host cell antioxidant response to regulate homeostasis and maintain survival of infected cell clones . We report that HBZ activates transcription of a set of oxidative stress response genes , among which is HMOX1 . Focusing on this gene , we found that HBZ-mediated transcriptional activation did not occur through Nrf2/small Maf heterodimers . Instead , our data support a model in which transcriptional activation occurs through the formation of HBZ/small Maf heterodimers that bind MAREs located in an HMOX1 enhancer region . Consistent with this mechanism , HTLV-1-infected T-cell lines were found to be more resistant to heme cytotoxicity than uninfected T-cell lines , and chemical inhibition of HMOX-1 reduced cell viability of HTLV-1-infected cells . Furthermore , knockdown of HBZ expression increased the oxidative state of HTLV-1-infected T-cells . This observation may be explained by the removal of a safeguard against ROS and RNS accumulation that results from the functions of Tax and p13 . Indeed , we found that HTLV-1-infected T-cells lines with active 5’ LTR transcription ( expressing Tax and p13 ) exhibit elevated oxidative states while an infected T-cell line with inactive 5’ LTR transcription does not . Overall , our findings identify a novel HBZ-dependent , pro-survival mechanism that may contribute to HTLV-1 persistence , facilitate successful transformation of infected cells , and help sustain ATL cells in the host .
HBZ is known to alter cellular gene expression by affecting the functions of transcriptional regulators; however , an understanding of the downstream physiological impact of these changes remains incomplete . To evaluate HBZ-induced changes in gene expression on a large scale , we previously performed a microarray analysis , comparing gene expression levels between a HeLa clonal cell line that stably expresses HBZ and a cell line containing the empty vector [44] . In manually annotating genes exhibiting potentially higher expression with HBZ , we identified a set of genes that are activated by Nrf2 and small Maf transcription factors under conditions of oxidative stress [3–6] . These genes included HMOX1 ( encodes one of the heme metabolizing enzymes ) , FTH1 ( encodes the heavy chain of the Ferritin iron transporter ) , and SQSTM1 ( encodes the stress-induced autophagy receptor , Sequestosome 1 ) . Additionally , we identified TNFRSF1A ( encodes a tumor necrosis factor α receptor ) and PIM1 ( encodes the proto-oncogene serine/threonine-protein kinase ) , which also appear to be activated by oxidative stress [4–6] . Published ChIP-Seq data support that Nrf2 and MafK bind in proximity to the transcribed regions of these five genes [4–6 , 45] . We verified the microarray results using quantitative , reverse-transcriptase PCR ( qRT-PCR ) to show that the mRNA levels of these oxidative stress-response genes were higher in HeLa cells stably expressing wild type ( WT ) HBZ than in cells containing the empty vector ( pcDNA ) ( Fig 1A ) . HBZ contains central basic regions that facilitate its nuclear import [46] and two transcriptional regulatory domains: an N-terminal activation domain ( AD ) and a C-terminal basic leucine zipper ( bZIP ) domain ( Fig 1B ) . The AD interacts with the cellular coactivators p300 and CBP via two LXXLL motifs [47 , 48] . The leucine zipper ( ZIP ) of the bZIP domain mediates dimerization with specific cellular bZIP transcription factors [33 , 49–52] . To determine which of these domains is important for transcriptional upregulation of the oxidative stress response genes , we performed a qRT-PCR analysis of HeLa clonal cell lines stably expressing HBZ with either LL→AA mutations in both LXXAA motifs of the AD ( HBZ MutAD ) or L→C mutations in the second and fourth heptad repeats of the ZIP domain ( HBZ MutZIP ) ( Fig 1B ) . Because the hbz mRNA has been reported to affect gene expression [35] , we also tested a cell line that does not express the HBZ protein due to an A→T mutation in the ATG start codon of the gene ( HBZ ΔATG ) . We found that expression of all five oxidative stress response genes was significantly reduced in the absence of the HBZ protein ( HBZ ΔATG ) and by mutations in either the AD or ZIP domain of HBZ ( Fig 1A ) . These results suggest that both transcriptional regulatory domains play roles in upregulating the expression of these genes . To better understand the mechanism through which oxidative stress response genes are upregulated in HBZ-expressing cells , we focused on HMOX1 , as transcriptional regulation of this gene and HMOX-1 enzymatic function are well-characterized [10 , 53] . This gene was also of interest based on its overexpression in certain cancers and on the association of HMOX-1 activity with survival and proliferation of tumor cells [11–13] . Through Western blot analysis , we first confirmed that the level of HMOX-1 was higher in HBZ-expressing HeLa cells than in cells containing the empty vector ( Fig 1C ) . We then evaluated the enzymatic activity of HMOX-1 using LC-MS to measure the conversion of heme into biliverdin in cell lysates ( Fig 1D ) . In these assays , hemin chloride served as the substrate . In addition , because biliverdin is rapidly converted to bilirubin by biliverdin reductase , bilirubin oxidase was added to facilitate its conversion back into biliverdin . Consistent with the qRT-PCR and Western blot data , cell lysates prepared from HBZ-expressing HeLa cells produced more biliverdin than lysates from cells expressing the HBZ mutants or containing the empty vector ( Fig 1E ) . These results show that elevated HMOX-1 levels in HBZ-expressing cells is accompanied by an increase in HMOX-1 enzyme activity . We next used qRT-PCR to measure levels of the HMOX1 transcript in a panel of uninfected and HTLV-1-infected T-cell lines . The uninfected cells that were tested included the acute T-cell lymphoblastic leukemia cell lines , Jurkat and CEM , as well as resting and anti-CD3/anti-CD28-activated primary CD4+ T-cells . The HTLV-1-infected T-cell lines that were tested included MT-2 , which was established in vitro , and ATL-2s and TL-Om1 , which were derived directly from ATL patients . TL-Om1 cells are unique among the HTLV-1-infected cell lines , as the 5’ LTR promoter in these cells is hypermethylated , which blocks expression of sense strand-encoded proteins , including Tax [54] . We compared levels of HMOX1 transcripts in the primary cells and cell lines to those in Jurkat cells ( Fig 2A ) . Using this approach , we observed a higher level of the transcript in resting , primary CD4+ T-cells , which was further elevated in the activated CD4+ T-cells . Given that HMOX-1 expression is upregulated during T-cell activation [55] , this result was expected . Interestingly , all three HTLV-1-infected T-cell lines , including the TL-Om1 cells that only express HBZ , exhibited transcript levels that were higher than those of the uninfected T-cell lines and the primary T-cells . Using Western blotting we confirmed that HMOX-1 protein levels are also higher in HTLV-1-infected cells than in uninfected cells ( Fig 2B , upper panels ) . In these assays we analyzed the membrane fractions from the different cell lines , as HMOX-1 is anchored in the membrane of the endoplasmic reticulum [56] . Voltage-dependent anion-selective channel protein ( VDAC ) was used as the loading control , as it is anchored to the mitochondrial membrane and concentrated in the membrane fraction [57] . We also analyzed HBZ levels in these cell lines using GST-KIX to enrich for HBZ from whole cell extracts ( Fig 2B , lower panels ) [58] . To evaluate the contribution of HBZ to HMOX-1 expression in HTLV-1-infected cell lines , we compared HMOX-1 protein levels between MT-2 cells expressing an shRNA that targets HBZ and MT-2 cells expressing a control shRNA that targets GFP ( Fig 2C , lower panels; S1A Fig ) [38 , 58] . Analysis of the membrane fractions showed that HMOX-1 protein levels were lower in the HBZ knockdown cells ( Fig 2C , upper panels ) . These results were supported by an in silico analysis of microarray data from ATL patient-derived cells ( herein denoted as ATL cells ) in which HBZ expression was knocked out [59] . In this former study , mRNA levels were quantified seven and eight days following induction of CRISPR/Cas9-mediated editing of hbz alleles . Both time points showed a reduction in HMOX1 mRNA ( S1B Fig ) . Consistent with our findings , these data indicate that reduced HBZ expression is associated with lower HMOX-1 expression . We next evaluated whether HMOX1 expression was elevated in PBMCs from patients presenting with an HTLV-1-associated disease . Using qRT-PCR , we compared HMOX1 transcript levels among CD8+ T-cell-depleted PBMCs collected from acute ATL patients ( n = 8 ) , HAM/TSP patients ( n = 8 ) , and asymptomatic HTLV-1 carriers ( n = 11 ) ( Fig 2D ) . Interestingly , we found HMOX1 transcript levels were significantly higher in both the ATL and HAM/TSP patient groups compared to asymptomatic carriers . However , we did not observe a significant correlation between HMOX1 transcript levels and the proviral load in the patient samples ( S2 Fig ) . To expand upon these patient results , we evaluated HMOX1 expression from published microarray data sets corresponding to another pool of HTLV-1-infected individuals [60 , 61] . In this former study , CD4+ T-cells were isolated from healthy donors , asymptomatic HTLV-1 carriers , and from patients with an indolent form of ATL ( chronic or smoldering ) or patients with acute ATL . By analyzing cell-surface expression of CADM1 and CD7 , the authors found that an increase in CADM1 and a decline in CD7 correlate with ATL progression . When we analyzed these data sets , we observed that HMOX1 transcript levels tend to increase from the CADM1neg CD7pos phenotype to the CADM1pos CD7neg phenotype , suggesting that overexpression of HMOX-1 may be related to ATL disease progression ( Fig 2E ) . Activation of the oxidative stress response is largely dependent on the transcriptional regulator , Nrf2 . During oxidative stress , transcription of the NFE2L2 gene , which encodes Nrf2 , increases [3] . In addition , Nrf2 , itself , is stabilized and accumulates in the nucleus [3] . We first evaluated whether HBZ-dependent upregulation of antioxidant genes is related to Nrf2 activity by analyzing NFE2L2 transcript levels in HBZ-expressing cell lines and in HTLV-1-infected T-cell lines . qRT-PCR analysis of NFE2L2 transcripts in HeLa cells stably expressing HBZ showed levels similar to those found in empty vector control cells ( Fig 3A ) . In addition , among the three HTLV-1-infected cell lines tested , only MT-2 cells exhibited a significantly higher NFE2L2 transcript level than Jurkat cells ( Fig 3B ) . Therefore , elevated HMOX-1 expression in infected cells and by HBZ alone does not correlate with upregulation of NFE2L2 gene transcription . In addition to increasing NFE2L2 gene transcription , oxidative stress also induces nuclear translocation of Nrf2 [3] . Under homeostatic conditions , Keap1 binds Nrf2 in the cytoplasm and , as an adaptor for a Cul3-dependent ubiquitin ligase complex , mediates proteasomal degradation of Nrf2 ( Fig 3C , left schematic ) . Without Nrf2 in the nucleus , Bach1 dimerizes with MafG ( or one of the other two small Mafs ) at MARE sites within antioxidant gene promoters , forming a complex that is transcriptionally repressive . During oxidative stress , oxidation of Keap1 disables the Nrf2 degradation process . In parallel , Nrf2 undergoes posttranslational modification that allows it to traffic to the nucleus . These events lead to the accumulation of Nrf2 in the nucleus , and through a separate set of posttranslational modifications , Bach1 is exported from the nucleus ( Fig 3C , right schematic ) . Given that Nrf2 contains an activation domain [62] , Nrf2/small Maf dimers bound to MAREs activate antioxidant gene transcription . We qualitatively compared the cytoplasmic/nuclear distribution of Nrf2 , MafG and Bach1 in uninfected and HTLV-1-infected T-cell lines ( Fig 3D ) . In these assays , MEK1/2 served as a marker for the cytoplasmic fraction [62] , and SP1 as a marker for the nuclear fraction as it is enriched in this fraction [63 , 64] . Surprisingly , we found that the overall levels of Nrf2 were lower in HTLV-1-infected cells , which was more pronounced when comparing the cytoplasmic fractions between the cell sets . These results suggest that Nrf2 undergoes more rapid turnover in infected cells and is , therefore , not mediating the increase in HMOX1 gene expression in these cells . Similarly , MafG was concentrated in the nucleus in both uninfected and HTLV-1-infected cells , which is consistent with previous reports [65 , 66] . Interestingly , when we analyzed the distribution of Bach1 , we found that it was highly enriched in the cytoplasmic fractions of the HTLV-1-infected cells , while conversely , it was more prevalent in the nuclear fraction of uninfected cells . These results suggested that , in HTLV-1-infected cells , transcription of antioxidant genes is upregulated through loss of Bach1-mediated repression , surprisingly , without involvement of Nrf2 . Given our evidence that HBZ upregulates HMOX1 and other oxidative stress-response genes , we tested whether HBZ alone effected a change in the cytoplasmic/nuclear distribution of Bach1 and Nrf2 . In comparing HBZ-expressing and empty vector HeLa clones , we did not observe a significant difference in the distribution of Bach1 between the cytoplasm and the nucleus ( Fig 3E; S3 Fig ) . This observation suggests that , in HeLa cells , HBZ alone is unable to further redistribute Bach1 to the cytoplasmic compartment . In addition , HBZ did not affect the cytoplasmic/nuclear distribution of Nrf2 ( Fig 3E; S3 Fig ) . Therefore , HBZ-mediated activation of HMOX-1 expression in these cells may not depend on nuclear localization of Nrf2 . Because HBZ did not appear to increase HMOX-1 expression through Nrf2 , we hypothesized that HBZ regulates antioxidant gene expression through direct interactions with the small Mafs . Indeed , the bZIP domain of HBZ was previously found to interact with MafG in vitro [67] . Moreover , we have used a proteomic approach to identify cellular proteins that interact with HBZ , and from this analysis , all three small Mafs arose as potential HBZ-binding partners ( Fig 4A ) [68] . The small Maf peptides identified are shown in S4 Fig . We confirmed these interactions using co-immunoprecipitation assays in which HEK 293T cells were first transfected with expression vectors for HBZ containing a C-terminal Myc-His tag and either MafG or MafK containing a C-terminal FLAG tag . Using antibodies against the epitope tags , HBZ was co-immunoprecipitated with both small Mafs from whole-cell extracts , and likewise , both small Mafs were co-immunoprecipitated with HBZ ( Fig 4B and 4C ) . The transiently-expressed small Mafs were detected as two bands by Western blot , potentially due to a second in-frame start site within the cDNA sequence . Additionally , we co-immunoprecipitated endogenous HBZ with endogenous MafG from whole-cell lysates prepared from ATL-2s cells ( Fig 4D ) , verifying that the HBZ/small Maf interaction occurs in ATL cells . As bZIP transcription factors , small Mafs form homo- and heterodimers with other bZIP factors through compatible leucine zipper ( ZIP ) interactions [69] . To confirm that the ZIP domain of HBZ mediates binding to the small Mafs , we performed co-immunoprecipitation assays from HEK 293T transfected with expression vectors for MafG and wild-type HBZ or HBZ MutZIP . As expected , wild-type HBZ interacted with MafG , but HBZ MutZIP did not ( Fig 4E ) , supporting that the interaction is mediated through the ZIP domains . A critical function of the small Mafs is to dimerize with antioxidant response regulators and thereby stably tether these factors to MAREs within antioxidant responsive gene promoters [70] . Many MAREs contain a core TPA-responsive element flanked by GC dinucleotides ( GC boxes ) as follows: 5’-TGCTGACTCAGCA-3’ ( the GC boxes are underlined , and the core is in bold ) [2 , 69] . Given the core sequence , this cis-element is denoted as a T-MARE [2] . The GC boxes , in addition to the core element , are critical for small Maf dimers to bind DNA . In the context of small Maf/Nrf2 heterodimers , the 3’ GC box is important for small Maf-binding and 5’ region of the MARE is generally recognized by Nrf2 [2] . Previously , Reinke et . al . provided biochemical evidence that dimers comprised of the bZIP domains of MafG and HBZ bind the T-MARE sequence [67] . Based on these data , we were interested in testing whether full-length HBZ is incorporated into a small Maf/MARE complex . To assess the DNA-binding activity of a full-length HBZ/small Maf complex , we performed EMSAs using a DNA probe containing the T-MARE sequence flanked by sequences identical to those found in the probes used by Reinke et . al . ( Fig 4F ) [67] . As a negative control , we used a DNA probe harboring substitutions in the GC boxes , which negatively impact small Maf-binding [71] . In addition to full-length HBZ , we tested the activation domain of HBZ ( HBZ-AD ) , which does not bind to the small Mafs . Prior to performing EMSAs , we used GST pull-down assays to verify that the recombinant , purified proteins used in these experiments exhibited their characteristic protein binding activities ( S5A Fig ) . As expected , we found that MafG interacts stably with the T-MARE probe but not with the mutant probe ( Fig 4F , lanes 2 and 4 ) . In the absence of MafG , both full-length HBZ and HBZ-AD failed to bind the T-MARE probe ( Fig 4F , lanes 7–8 ) . However , the combination of full-length HBZ and MafG substantially increased formation of the T-MARE-bound protein complex over that of MafG alone ( Fig 4F , lanes 6 and 10 ) . In contrast , HBZ-AD did not alter formation of the MafG/T-MARE complex ( Fig 4F , lanes 6 and 9 ) . These data are consistent with previously published observations [67] , supporting a mechanism in which HBZ forms heterodimers with the small Mafs that are capable of binding the T-MARE sequence . To provide support that HBZ is indeed incorporated into the protein complex , we used immobilized DNA-binding assays in which the T-MARE probe was biotinylated and coupled to streptavidin beads that were then incubated with nuclear extracts . The nuclear extracts used were prepared from HEK 293T cells transfected with individual or both expression plasmids for MafG-FLAG and HBZ-Myc-His . As expected , MafG-FLAG bound the MARE probe independently of HBZ ( Fig 4G , lane 4 ) . In the absence of MafG-FLAG , we did not detect binding of HBZ to the T-MARE probe ( Fig 4G , lane 5 ) , possibly due to competition with Nrf2 and Bach1 for binding to the endogenous small Mafs . However , when MafG-FLAG was also present in extracts , HBZ did bind to the probe ( Fig 4G , lane 6 ) . Neither HBZ nor MafG-FLAG bound to DNA-free beads ( Fig 4G , lanes 7–9 ) , suggesting that binding to the T-MARE probe was specific . Using the immobilized DNA-binding assay with recombinant , purified proteins produced similar results ( S5B Fig ) . These data further support the mechanism of HBZ-recruitment to T-MAREs through interactions with the small Mafs . Given the important role of the small Mafs in regulating antioxidant gene transcription , we hypothesized that HBZ , through interactions with the small Mafs , directly activated HMOX1 transcription . ChIP-Seq data available through the UCSC Genome Browser [45] for the MafK revealed two regions of MafK enrichment upstream of HMOX1 . These regions were originally defined as enhancers [62] , and we refer to them as the distal and proximal binding peaks ( Fig 5A ) . Sequence analysis revealed that the distal peak contains three MAREs ( Distal 1–3 ) , while the proximal peak contains a single MARE ( S6A Fig ) . To test whether HBZ associates with these peak regions , we performed ChIP assays using HeLa clones that stably express HBZ-Myc-His or the translation-deficient mutant of HBZ ( HBZ ΔATG ) , or contain the empty expression vector ( pcDNA ) . We probed for HBZ using an antibody against its His-epitope tag , and additionally , we probed for MafG and Nrf2 using protein-specific antibodies . qPCR analysis of immunoprecipitated DNA examined the peak regions and a downstream negative control region located within the HMOX1 gene ( Fig 5A ) . As expected , we found that MafG was highly enriched at both peak regions in all three cell lines ( Fig 5B ) , supporting the concept that the small Mafs may serve as “stand-in” factors that help poise the chromatin for transcription [72] . In contrast to the MafG pattern , significant enrichment of Nrf2 was only observed at the distal peak region with the exception that , in HBZ-expressing cells , Nrf2 was not significantly enriched at this region ( Fig 5C ) . Interestingly , we observed significant enrichment of HBZ at the distal peak region ( Fig 5D ) , suggesting that HBZ increases HMOX-1 expression through association with HMOX1 gene promoter . We were also interested in determining whether HBZ binds to the same region of the HMOX1 promoter in HTLV-1-infected T-cells . Because we do not have a ChIP-grade antibody against HBZ , we analyzed published ChIP-Seq data from experiments designed to identify HBZ-binding sites in the ATL cell line , KK1 [59] . Interestingly , a peak of HBZ-enrichment in the ATL cells overlaps with the distal peak of MafK-enrichment in HeLa cells ( Fig 5A ) . Further analysis of the peak sequences confirmed this observation ( S6B Fig ) and is consistent with our ChIP results in HeLa cells , providing evidence that HBZ is a direct regulator of HMOX1 transcription in ATL cells . In conjunction with this analysis we analyzed MafG-binding to the HMOX1 promoter in TL-Om1 cells . Strikingly , we observed significant enrichment of MafG at the distal peak region where HBZ was found to be enriched according to the in silico analysis ( Fig 5E ) . These data suggest that HBZ associates with the HMOX1 promoter through interactions with the small Mafs . The distal peak region in the HMOX1 promoter contains three separate MAREs ( Fig 5A ) . To determine which of these is bound by HBZ/small Mafs heterodimers , we performed EMSAs using probes that encompass each of the three MAREs with 10 or 11 base-pairs of its flanking genomic sequence . These probes were designated Distal 1 , Distal 2 , and Distal 3 ( S6A Fig ) . In addition , we analyzed binding to a probe encompassing the proximal MARE . To test protein interactions with these probes , we used recombinant , purified MafG and HBZ-bZIP . The bZIP domain of HBZ alone was used in these experiments to further verify that this domain was sufficient for formation of the DNA-bound complex . Importantly , we found that , like the full-length protein , the bZIP domain of HBZ increased formation of the T-MARE/protein complex ( Fig 6 , lanes 1–6 ) . When we tested binding to the new probes , we found that MafG alone appeared to exhibit a higher affinity for the proximal MARE sequence than for the three distal MARE sequences ( Fig 6 , lanes 8 , 11 , 14 and 17 ) . Interestingly , in the case of the three distal MARE probes , HBZ-bZIP increased formation of the protein/DNA complex , an effect that was most pronounced with the Distal 3 probe ( Fig 6 , lanes 9 , 12 and 15 ) . In contrast , HBZ-bZIP diminished formation of a complex with the proximal MARE probe , indicating that HBZ/small Maf heterodimers do not bind the proximal MARE . Overall , these results suggest that HBZ/small Maf heterodimers bind the distal MAREs in the HMOX1 promoter . To evaluate the transcriptional activity of HBZ from MAREs , we constructed a luciferase reporter plasmid with four tandem repeat sequences containing the T-MARE , which were inserted upstream of a minimal promoter ( 4xT-MARE; Fig 7A ) . We then performed reporter assays using Jurkat cells co-transfected with the 4xT-MARE reporter plasmid , or the reporter plasmid containing only the minimal promoter ( minP ) , as well as increasing quantities of an expression vector for wild-type HBZ . Results from these assays showed that HBZ significantly activates transcription from the T-MARES ( Fig 7B ) . In contrast , HBZ MutZIP had no effect on transcription from these sites ( Fig 7C ) . These data show that HBZ activates transcription from T-MAREs , and support a model that it does so through interactions with the small Mafs . To verify that this transactivation by HBZ depends on the small Mafs , we performed similar luciferase reporter assays with the addition of an Nrf2 dominant negative mutant ( Nrf2-DN ) . This mutant harbors an N-terminal deletion that removes its Keap1-binding site ( Neh2 ) and activation domain ( Neh4 , Neh5 ) , but retains its CNC-bZIP domain ( Fig 7D ) [62] . Therefore , Nrf2-DN competes with other small Maf-binding partners , forming transactivation-defective heterodimers with the small Mafs on MAREs . To assess the effects of this mutant on transactivation from the T-MAREs , we co-transfected Jurkat cells with the 4xT-MARE or the minP reporter plasmid and expression vectors for wild-type HBZ and/or Nrf2-DN ( Fig 7E ) . We observed that Nrf2-DN alone slightly increased 4xT-MARE transcription , which may be a de-repressive effect caused by competition between the mutant and Bach1 for small Maf-binding ( Fig 7E , lane 3 ) . As expected , HBZ activated 4xT-MARE transcription ( Fig 7E , lane 2 ) ; however , this effect was significantly diminished in the presence of Nrf2-DN ( Fig 7E , lane 4 ) and was also found to be dose-dependent ( Fig 7F ) . Consistent with these results , ectopic expression of Nrf2-DN in HeLa cells stably expressing HBZ significantly reduced levels of the endogenous HMOX1 transcript ( Fig 7G ) . These data further support the model that HBZ upregulates transcription from MAREs by forming dimers with the small Mafs . Though small Mafs are critical regulators of antioxidant gene expression , they lack activation domains . Therefore , gene expression is upregulated by small Mafs when they form MARE-bound heterodimers with Nrf2 , which provides these complexes with an activation domain [73 , 74] . Given that the activation domain of HBZ functions through interactions with the paralogous coactivators , p300 and CBP [47 , 48] , we hypothesized that HBZ acts similarly to Nrf2 when bound to the small Mafs . Indeed , as shown in Fig 1A , in comparison to wild-type HBZ , HBZ MutAD exhibited a significant decrease in its ability to activate antioxidant gene expression . Consequently , we tested this hypothesis using co-immunoprecipitation assays to evaluate binding of the coactivators to HBZ/small Maf complexes . HEK 293T cells were transfected with expression vectors for MafG and/or HBZ , and proteins from cell lysates were immunoprecipitated using a MafG antibody . From these assays , we found that p300 and CBP only co-immunoprecipitate with MafG when HBZ is present ( Fig 7H ) , suggesting that HBZ activates antioxidant gene transcription by recruiting p300/CBP to the promoters of these genes . Previous reports have shown that expression of the HTLV-1-encoded proteins , Tax and p13 , results in the accumulation of ROS/RNS [21 , 22 , 24–27 , 75] . Since we found that HBZ upregulates expression of HMOX1 and other antioxidant genes , we wanted to test whether HBZ protects HTLV-1-infected cells from Tax- or p13-induced oxidation . To examine this possibility , we first measured the ratios of reduced to oxidized glutathione in uninfected and HTLV-1-infected T-cells as a means of assessing the redox states of these cells . Glutathione is a ubiquitous , non-enzymatic antioxidant that , in its reduced form ( GSH ) , is a tripeptide with a single cysteine residue [76] . Upon exposure to free radicals , the thiol group of this cysteine is oxidized , resulting in glutathione disulfide ( GSSG ) ( Fig 8A ) . In experiments , LC-MS was used to quantify levels of free GSH and GSSG in each cell lines which is reported as a ratio ( GSH:GSSG ) . From this analysis , we found that Jurkat and CEM cells exhibited GSH:GSSG ratios that are within the range reported for normal , unstressed cells ( Fig 8B ) , which for healthy cells , is 100:1 or greater [76–78] . Among the HTLV-1-infected cell lines , ATL-2s and MT-2 cells had significantly lower GSH:GSSG ratios indicative of a more oxidized cellular state . In contrast , TL-Om1 cells , which do not express Tax or p13 , exhibited a redox state similar to those of the uninfected cells . This general pattern is consistent with the pro-oxidative effect of the two viral proteins . We then used the same approach to compare the redox states between MT-2 cells expressing an shRNA that targets HBZ and MT-2 cells expressing an shRNA that targets GFP . We found that the GSH:GSSG ratio was significantly lower in the cells expressing the shRNA that targets HBZ ( Fig 8C ) , indicating that HBZ helps prevent the accumulation of ROS/RNS in HTLV-1-infected cells . Constitutive HMOX-1 expression has been demonstrated to confer resistance to cellular stressors , including its substrate heme , as well as to some chemotherapeutic agents [11–13] . Many HTLV-1-infected T-cell lines , including patient-derived ATL cell lines , exhibit increased resistance to a variety of stress-inducing chemotherapy agents , including cisplatin , doxorubicin , and etoposide [79–85] . We hypothesized that the anti-oxidative effect of HBZ is cytoprotective and promotes cell survival during oxidative stress . We used HeLa cells either stably expressing HBZ or carrying the empty vector to test this hypothesis . Cell-viability in each cell line was assessed using MTT assays after challenging cells with hemin to induce iron-mediated oxidative stress . We found that , following hemin-treatment , cells carrying the empty vector exhibited a significant reduction in viability , while HBZ-expressing cells were unaffected by the treatment ( Fig 9A ) . To expand upon these results , we used alamarBlue cell viability assays to compare the effects of hemin cytotoxicity between uninfected ( Jurkat and CEM ) and HTLV-1-infected T-cells ( TL-Om1 , ATL-2s and MT-2 ) . We found that exposure of Jurkat and CEM cells to varying concentrations of hemin cause substantial cell-death , while in comparison , these treatments were significantly less effective at killing the HTLV-1-infected T-cells ( Fig 9B ) . To test whether the hemin-resistance of the infected cells is dependent on HMOX-1 , we used a small molecule inhibitor of HMOX-1 , 2-[2- ( 4-bromophenyl ) ethyl]-2-[ ( 1H-imidazol-1-yl ) methyl]-1 , 3-dioxolane hydrochloride ( OB-24 ) [86] . In these experiments , cells were treated with OB-24 and hemin in combination or alone . For uninfected cells , the combination of the two compounds did not decrease cell viability below that produced by hemin alone ( Fig 9C ) . Given our initial findings that these cells are already highly susceptible to hemin-mediated cytotoxicity , this result was expected . In contrast , all three HTLV-1 infected cell lines exhibited a significantly reduction in cell survival with the two compounds together compared to hemin alone ( Fig 9C ) . These results show that the increased expression of HMOX-1 in HTLV-1-infected cells helps avert oxidative stress , and thereby promotes the survival of these cells .
In this study , we found that HBZ activates the expression of a group of antioxidant genes that are normally induced by oxidative stress . Using the HMOX1 gene as a model , we provide evidence that HBZ activates transcription of this gene by forming heterodimers with small Mafs ( MafF , MafG or MafK ) at MARE sites located in an upstream enhancer . As the small Mafs lack activation domains , they rely on interacting partners , such as Nrf2 , to activate transcription [2] . We provide evidence that such a mechanism applies to small Maf/HBZ heterodimers , with HBZ supplying an activation domain that mediates recruitment of the coactivator , p300/CBP , which in turn leads to activation of transcription ( Fig 10 ) . This mechanism is remarkable considering that the basic region of the bZIP domain of HBZ lacks consensus amino acid motifs found in other bZIP factors . Given the critical role of the basic region in DNA-binding , heterodimers formed by HBZ and one of a variety of other cellular bZIP factors often fail to bind DNA , and in this context , HBZ functions as a transcriptional repressor [33 , 49 , 51 , 52] . In lieu of this general mechanism , Reinke et al . previously provided strong biochemical evidence that heterodimers composed of the bZIP domains of MafG and HBZ exhibit DNA-binding activity [67] . There are multiple ARE subtypes , and in this former study , a T-MARE was used , which consists of a central TRE flanked by GC-boxes [2] . The GC boxes are known to be important for small Maf-binding , and interestingly , according to our results , they may also be important for HBZ-binding . Indeed , our data indicate that HBZ/small Maf heterodimers bind the distal HMOX1 enhancer AREs ( denoted as MAREs here ) that contain both 5’ and 3’ GC boxes , but not the proximal ARE that lacks a 3’ GC box . Because substantial sequence variation exists among the AREs , future studies will be needed to clarify which cis-acting elements are targeted by HBZ/small Maf heterodimers and , more specifically , the consensus motif recognized by these complexes . It should be noted that HBZ has also been reported to interact with one of the large Mafs , MafB , which contrary to our observations with small Mafs , inhibits DNA-binding and transcription [87] . While large and small Mafs exhibit high sequence similarity ( S4 Fig ) , large Mafs are distinct in that they contain an activation domain and display tissue-specific expression [88] . In the previous study , the DNA sequence used to assess DNA-binding lacked a 5’ GC box , which may have impaired binding by a MafB/HBZ complex . Interestingly , in the study by Reinke et al . , heterodimers composed of the bZIP domains of MafB and HBZ did appear to bind the T-MARE sequence , suggesting that HBZ/MafB heterodimers do in fact bind certain AREs [67] . However , the relevance of MafB/HBZ heterodimers will depend on whether MafB is expressed in HTLV-1-infected T-cells , which has not yet been assessed . In EMSAs HBZ appears to increase the stability of the protein complex formed on ARE sequences . This observation is reminiscent of the effect of the viral protein Tax on CREB in the context of the Tax-responsive element-1 ( TxRE1 ) sites that regulate transcription of the HTLV-1 provirus [19] . In the absence of Tax , CREB binds weakly to these sequences; however , when Tax associates with CREB , TxRE1-bound protein complexes are substantially more stable [89 , 90] . This effect of Tax is instrumental to transcriptional activation of the HTLV-1 provirus . In a similar manner stabilization of ARE-bound protein complexes by HBZ may be significant to the mechanism by which HBZ activates transcription from AREs . Among the antioxidant genes found to be regulated by HBZ , we focused primarily on characterizing downstream effects of increased HMOX1 expression in relation to HTLV-1 infection and ATL . While HMOX-1 expression is induced by oxidative stress in healthy cells to serve a beneficial and tumor-suppressive role [10] , unregulated expression of HMOX-1 acquired by certain cancers is linked to increased cell-survival and proliferation [11–13] . Additionally , HMOX-1 expression has been reported to be induced in response to chemotherapies and radiotherapies , which may help confer cancer cells with multi-drug resistance [11–13] . These adverse effects of HMOX-1 coincide with features of ATL , as ATL patients do not typically respond well to chemotherapeutic regimens , and disease progression and relapse after treatment are often associated with the onset of multi-drug resistance [91] . Moreover , ATL cells and HTLV-1-transformed T-cell lines in culture display resistance to many clinically-relevant anti-cancer drugs [79–85] . In line with these observations , we found that HMOX-1 expression is elevated in HTLV-1-infected T-cell lines and in PBMCs from a set of patients with acute ATL . Furthermore , an analysis of published microarray data [60 , 61] suggests that increasing HMOX1 expression is associated with the worsening of ATL symptoms . To date , the drug resistance of ATL cells has been associated with viral-mediated activation of pro-survival mechanisms that bypass cell-cycle regulatory checkpoints , inhibit the induction of apoptosis , and increase drug efflux from the cells [79–85] . In this study we identified an additional mechanism that involves resistance to ROS-induced heme cytotoxicity ( Fig 10 ) . Heme is released by hemoproteins that are damaged through the effect of ROS . The resulting free heme can then promote peroxidation of membrane lipids , protein fragmentation and DNA damage , ultimately leading to cell death [92] . Through degradation of free heme by HMOX-1 , these effects are averted . Indeed , we found that , when challenged with heme , the elevated expression of HMOX-1 in HTLV-1-infected T-cells increased cell-survival . Interestingly , multiple approaches have been used to reduce HMOX-1 expression in different types of cancer , which have , overall , produced a variety of positive effects , such as increases in sensitivity to anticancer drug-induced apoptosis and reductions in proliferation and invasiveness [13] . Based on these observations , the development of clinical inhibitors of HMOX-1 would likely benefit ATL patients by improving the effectiveness of the current chemotherapeutic regimens used to target the malignant cells . In addition to possibly impeding anticancer drug-effects , an HBZ-mediated increase in HMOX-1 abundance in infected cells may counteract cytotoxic effects caused by other HTLV-1-encoded proteins . The viral protein , Tax , is essential for HTLV-1 replication , as it activates transcription from the 5’ LTR of the provirus [19] and stimulates mitotic expansion of infected cells [93] . This latter function occurs through the dysregulation of a variety of cellular pathways by Tax . While the apparent goal of this process is to promote cell proliferation , it also culminates in the accumulation of ROS/RNS [23 , 24] . This effect of Tax has been attributed to inhibition of stress granule formation and constitutive activation of NF-κB signaling , which activates iNOS expression [25 , 26] . Paradoxically , ROS generated through the effort to drive mitotic expansion of infected cells can trigger apoptosis or cellular senescence [41 , 94] . In connection with these outcomes and our current results , HBZ has been shown to offset Tax-mediated cellular senescence [41 , 42] . While attributed to a reduction in NF-κB signaling , this effect may also involve increased expression of HMOX-1 and other antioxidant proteins by HBZ that serve to detoxify Tax-mediated ROS/RNS . Consistent with this premise , HTLV-1-infected T-cell lines with Tax expression ( MT-2 and ATL-2s ) exhibited low GSH:GSSG ratios , indicative of oxidative stress . In contrast , TL-Om1 cells , which lack Tax expression due to transcriptional repression of the 5’ LTR promoter , displayed a GSH:GSSG ratio consistent with low oxidative stress . More importantly , knocking down HBZ expression in MT-2 cells resulted in a significant decline in oxidative stress , supporting that HBZ plays an important role in ameliorating the cell’s oxidative state . The HTLV-1-encoded protein , p13 , also stimulates ROS production . p13 functions by localizing to the inner mitochondrial membrane where it induces K+ influx , leading to mitochondrial swelling and depolarization , and , in turn , increased ROS production . Interestingly , transformed T-cells appear to be more sensitive to the cytotoxic effects caused by these events than primary T-cells , which has led to the proposal that p13 serves to eliminate HTLV-1-infectected T-cells that become transformed as a means of supporting long-term viral persistence in the host . In the context of this model , it is possible that HBZ usurps the role of p13 , thereby reinforcing the survival of virally-infected cells that have undergone transformation . Similar to our findings , another HTLV-1-encoded protein , p30 , was recently reported to suppress the cytotoxic effects of ROS [95] . This function originates from the ability of p30 to upregulate TP53-induced glycolysis and apoptosis regulator ( TIGAR ) [96] ) . TIGAR is a fructose-2 , 6-bisphosphatase that supports metabolism through the pentose phosphate pathway , leading to the production NADPH , which can be used to regenerate GSH from GSSG [97] . Interestingly , like HMOX-1 , TIGAR is implicated in cancer cell-survival and proliferation [97] . It is important to note that in this previous study , HBZ , in addition to Tax , was implicated in the accumulation of ROS . This discrepancy with our study may be due to the different methods used to measure oxidative stress . While Hutchison et al . used the JC-1 dye , which measures mitochondrial membrane potential [95] , we used mass spectrometry to quantify GSH:GSSG ratios . Despite the difference between studies , it is interesting that HTLV-1 appears to have adapted separate mechanisms to combat the cytotoxic effects of ROS . It is possible that these mechanisms evolved explicitly to support viral replication through Tax-driven mitotic expansion . However , they may also produce an unintended effect of supporting survival of infected cells that have undergone transformation .
The following mammalian expression plasmids have been described: pcDNA-HBZ Sp1-Myc-His ( aa 1–206 ) [34] , pcDNA-HBZ-ΔAD-Myc-His ( aa 77–206 ) [50] , pcDNA-HBZ-ΔbZIP-Myc-His ( aa 1–130 ) and pcDNA-HBZ-ΔATG [51] , pcDNA-HBZ- ( LXXAA ) 2-Myc-His [47] , pcDNA-HBZ-MutZIP-Myc-His and pSG-HBZ-Myc [44] , pCMV-HBZ-FLAG [40] , and pSG-Tax-His [98] . Empty vector plasmids pCMV-3Tag-8 and pSG5 were purchased from Agilent Technologies and pcDNA3 . 1 ( + ) / Myc-His A was purchased from Invitrogen . The HBZ-MutZIP-Myc sequence from pcDNA-HBZ-MutZIP-Myc-His was inserted into the EcoRI site of pSG5 to generate pSG-HBZ-MutZIP-Myc . The small Maf sequences from pDNR-Dual vectors ( DNASU plasmid repository: HSCD00002293 , HSCD00005183 and HSCD00004984 ) [99] were inserted between the BamHI and HindIII sites in pCMV-3Tag-8 to generate pCMV-MafK-FLAG and pCMV-MafG-FLAG . The sequence corresponding to amino acids 401–606 of Nrf2 from pcDNA3-Myc3-Nrf2 ( a gift from Yue Xiong; Addgene plasmid #21555 ) [100] was inserted between the BamHI and XbaI sites in pcDNA3 . 1 ( + ) / Myc-His A to generate the Nrf2 dominant negative expression plasmid , pcDNA-Nrf2-DN-Myc-His . The 4xT-MARE sequence ( GeneArt Gene Synthesis , Invitrogen ) 5’-CTCGAGTCGAGCTCGGAATTGCTGACTCAGCATTACTCTCGTCGAGCTCGGAATTGCTGACTCAGCATTACTCTCGTCGAGCTCGGAATTGCTGACTCAGCATTACTCTCGTCGAGCTCGGAATTGCTGACTCAGCATTACTCTCGGATCCAAGCTT-3’ was inserted between the XhoI and HindIII sites in pGL4 . 26 ( Promega ) to generate pGL-4xT-MARE-Luc . The bacterial expression plasmids pGEX-HBZ and pGEX-KIX have been described [47] . pGEX-4T-2 , pRSET A , and pET3A were purchased from GE Healthcare , Thermo Fisher Scientific , and Novagen , respectively . The sequence corresponding to amino acids 1–57 of HBZ from pcDNA-HBZ-Myc-His was inserted between the BamHI and EcoRI sites in pGEX-4T-2 to generate pGEX-HBZ-AD . The small Maf sequences from the pDNR-Dual vectors ( above ) were inserted between the BamHI and HindIII sites in pRSET A to generate pRSET A-MafF-His and pRSET A-MafG-His . The sequence corresponding to amino acids 120–206 from pcDNA-HBZ-Myc-His was inserted into the BamHI site in pET3A to generate T7-HBZ-bZIP . Newly constructed plasmids were sequenced . The following antibodies were used: anti-GST Tag ( G7781 ) , anti-Actin clone C4 ( MAB1501 ) , anti-FLAG M2 ( F3165 ) and anti-Myc clone 4A6 ( 05–724 ) were purchased from Millipore-Sigma; anti-His H-15 ( sc-803 ) , anti-p300 N15 ( sc-584 ) , anti-CBP A22 ( sc-369 ) , anti-Nrf2 C20 ( sc-722 ) , anti-Bach1 F9 ( sc-271211 ) and anti-HMOX1 A-3 ( sc-136960 ) were purchased from Santa Cruz Biotechnology; anti-VDAC D73D12 ( #12454 ) , anti-MEK1/2 L38C12 ( #4694 ) and anti-HMOX1 ( #70081 ) were purchased from Cell Signaling Technology; anti-MafG ( ab154318 and ab86524 ) and anti-6x His ( ab9108 ) were purchased from Abcam; hybridoma anti-MafG were purchased from DSHB ( Cat# PCRP-MAFG-1H7 , RRID:AB_2618829 ) ; and anti-SP1 ( 21962-1-AP ) was purchased from Proteintech . Anti-HBZ serum was a gift from Dr . Mesnard [33] . HEK 293T/17 ( ATCC , ATCC CRL-11268 ) were cultured in DMEM supplemented with 10% Fetalplex animal serum complex ( Gemini Bio-Products ) , 2 mM L-glutamine , 100 U/ml penicillin , and 50 μg/mL streptomycin . Clonal HeLa cell lines derived from HeLa-S3 ( a gift from Dr . Nyborg ) expressing HBZ-Myc-His , HBZ-ΔbZIP-Myc-His , HBZ-MutZIP-Myc-His , HBZ- ( LXXAA ) 2-Myc-His , and HBZ-ΔATG were cultured in supplemented DMEM maintained under selection with 0 . 5 mg/mL geneticin ( Thermo Fisher Scientific ) [44 , 68] . Jurkat and CEM cells ( a gift from Dr . Nyborg ) , MT-2 cells ( obtained from the NIH AIDS Research Program , #237 ) and ATL-2s cells ( a gift from Dr . Matsuoka ) were maintained in supplemented IMDM . TL-Om1 cells ( a gift from Dr . Matsuoka ) were maintained in supplemented RPMI . MT-2 cells stably expressing an shRNA that targets HBZ-SP1 ( V4 ) ( a gift from Dr . Green ) were maintained under selection with 1 mg/mL geneticin [38] . MT-2 cells stably expressing an shRNA against GFP ( MISSION pLKO . 1-puro eGFP shRNA , Thermo Fisher Scientific SHC005 ) , were established and maintained under selection in 1 . 5 μg/mL puromycin . Blood samples from symptomatic and asymptomatic HTLV-1-infected donors were obtained from the CHU of Martinique and isolated as previously described [58] . Clinical sample collections for research purposes are stored at the Center of Biological Resources of Martinique ( CeRBiM ) . We received approval from the CeRBiM Review Board to use the samples . HTLV-1 AC and patients suffering from TSP/HAM or ATL were recruited according to World Health Organization ( WHO ) criteria . AC had no neurologic or haematological symptoms . According to the French Bioethics laws , the collection of samples from AC , TSP/HAM and ATL patients has been declared to the French Ministry of Research . Because the protocol is non-interventional , no informed consent was required , as stated by the French Public Health code and , therefore , the study was conducted anonymously . Cells were equalized , cultured overnight , and RNA was extracted using TRIzol Reagent ( Invitrogen ) according to the manufacturer’s instructions . For assays using the Nrf2-DN mutant , 1 x 107 HeLa cells stably expressing HBZ-Myc-His cells were electroporated with 3 . 6 μg pMACS CD4 and 16 . 4 μg pcDNA or 16 . 4 μg pcDNA-Nrf2-DN-Myc , and positively transfected cells were enriched using the MACSelect System ( Miltenyi Biotec ) prior to RNA extraction as described [44] . cDNA was synthesized with random hexamers or oligod ( T ) primers using the RevertAid cDNA synthesis kit ( Thermo Fisher Scientific ) . Quantitative real-time PCR ( qRT-PCR ) amplification of cDNA was performed as described [44] . For analysis of diluted cDNA , standard curves were generated for each primer pair using serial dilutions of an appropriate experimental sample . PCR efficiencies from all plates and primer pairs ranged from 89 . 1% to 134 . 4% , with correlation coefficients >0 . 95 . Undiluted cDNA was used for analysis of HMOX1 mRNA in T-cell lines . Primer sequences are as follows: HMOX-1 , 5’-TGATAGAAGAGGCCAAGACTGCGT-3’ and 5’-TCGCCACCAGAAAGCTGAGTGTAA-3’; FTH1 , 5’-CGCCTCCTACGTTTACCTGT-3’ and 5’-AGCATGTTCCCTCTCCTCAT-3’; SQSRM1 , 5’-TTCTTTTCCCTCCGTGCTC-3’ and 5’-GGATCCGAGTGTGAATTTCC-3’; TNFRSF1A , 5’-ACGAGTGTGTCTCCTGTAGTAGTA-3’ and 5’- AACCAATGAAGAGGAGGGATAAA-3’; PIM1 , 5’-TTCTTCAGGCAGAGGGTCTCTTCA-3’ and 5’-TGTGGAGGTGGATCTCAGCAGTTT-3’; Nrf2 , 5’-CCAGCACATCCAGTCAGAAA-3’ and 5’-GACTGAAACGTAGCCGAAGAA-3’; UBE2D2 ( housekeeping gene ) , 5’-TGCCTGAGATTGCTCGGATCTACA-3’ and 5’-ACTTCTGAGTCCATTCCCGAGCTA-3’ . For assays using patient-derived cells , sample processing and qRT-PCR were performed as described [101] . PBMCs were depleted of the CD8+-T cells to prevent CD8+ T cell-mediated killing of the HTLV-1-infected cells in the specimens . Cells were cultured for 5 days prior to being analyzed . HMOX-1 enzymatic activity of was determined as described , but with modifications to accommodate the use of cultured cells [102] . Cells ( 2 x 106 ) were treated with 200 μM hydrogen peroxide in low serum media ( 0 . 5% Fetalplex ) and for 4h to stimulate HMOX-1 activity and then harvested by centrifugation at 4°C , washed with cold PBS , and suspended in 200 μL homogenization buffer ( 20 mM Tris-HCl [pH 7 . 5] , 250 mM sucrose , 1 mM EDTA ) supplemented with protease inhibitor cocktail ( Millipore-Sigma , P8340 ) . Cells were homogenized using a Dounce tissue grinder , and lysates containing 200 μg of protein were adjusted to 20 mM Tris-HCl [pH 7 . 5] , 250 mM sucrose , 12 . 5 μM hemin chloride ( Millipore-Sigma , 3741 ) , 1 mM NADPH ( Millipore-Sigma , 481973 ) , and 0 . 025 U bilirubin oxidase ( Millipore-Sigma , B0390 ) in equal volumes . Reactions were incubated at 37°C for 30 minutes , halted by adding an equal volume of 0 . 1% formic acid in methanol , and supplemented with 0 . 1 ng/μL Biliverdin d4 ( an internal standard; Millipore-Sigma , 795089 ) . Samples were then centrifuged at 15 , 000 x g at 4°C for 20 minutes , and clarified supernatants were analyzed by liquid chromatography mass spectrometry ( LC-MS ) using an Agilent 1200 series high performance liquid chromatograph connected to an Agilent 6220 time-of-flight ( TOF ) mass spectrometer . Chromatographic separation was performed using an Agilent Zorbax Eclipse Plus C18 column ( 3 . 5 μm , 2 . 1 × 150 mm ) held at 35°C . Mobile phase A consisted of water with 1% formic acid whereas mobile phase B consisted of acetonitrile with 1% formic acid . Flow rate was set to 0 . 25 mL/min . Initial solvent composition was 10% B which was held for 1 minute , ramped to 55% B over 1 minute , ramped to 90% B over the next 2 minutes , ramped to 100% B in 2 minutes , and held at 100% B for the next 5 minutes resulting in a total analysis time of 11 minutes . The TOF was operated in positive mode and biliverdin along with biliverdin-d4 were quantified using the [M+H]+ ion . Extracted [M+H]+ ion chromatograms were integrated to give biliverdin peak areas which were then normalized by the peak area of biliverdin-d4 . Biliverdin/biliverdin-d4 ratios were compared between groups to determine relative biliverdin concentrations . Whole cell extracts were prepared as described [68] using RIPA buffer ( 50 mM Tris [pH 8 . 0] , 1% Triton X-100 , 100 mM NaCl , 1 mM MgCl2 , 400 nM TSA , 2 μg/mL leupeptin , 5 μg/mL aprotinin , 1 mM phenylmethylsulfonyl fluoride [PMSF] , and 1 mM benzamidine ) . Isolation of membrane fractions was performed as described [57] . Nuclear and cytoplasmic fractions were prepared as described with slight modification [103] . Briefly , 1 . 5 x 106 PBS-washed cells were suspended in hypotonic buffer ( 20mM HEPES pH 7 . 9 , 20% [vol/vol] glycerol , 10 mM NaCl , 1 . 5 mM MgCl2 , 0 . 2 mM EDTA , 1 mM DTT , 0 . 1% NP40 , 2 μg/mL leupeptin , 5 μg/mL aprotinin , 1 mM PMSF , 1 mM benzamidine ) and ice-chilled for 10 minutes . Samples were then centrifuged at 0 . 6 x g and 4°C for 5 minutes . Supernatants ( cytoplasmic fraction ) were collected , and nuclei were lysed by adding 50 μL of RIPA buffer , vortexing and ice-chilling samples for 15 minutes . Lysates were centrifuged at 16 , 000 x g and 4°C for 15 minutes , and supernatants ( nuclear fractions ) were collected . Protein concentrations were determined by Bradford assays ( Bio-Rad ) . Co-IP assays were performed as described [68] . For assays with HEK 293T cells , 2 x 106 cells were transfected using TurboFect ( Thermo Fisher Scientific ) according to the manufacturer’s instructions ( plasmid amounts stated in the figure legend ) . Whole cell extracts were prepared from cells 24 to 48 hours after transfection , and 300 μg of extract was combined with anti-FLAG M2 magnetic beads ( Millipore-Sigma , M8823 ) or protein G agarose beads ( Millipore-Sigma , P7700 ) pre-bound with 2 μg of anti-Myc antibody . For assays with T-cells , 2 mg of whole cell extract were combined with protein G agarose beads prebound with an anti-MafG hybridoma supernatant ( DHSB ) or normal mouse serum ( Jackson ImmunoResearch laboratories , Inc . ) . SDS-PAGE , Western blot analysis and nitrocellulose membrane imaging were performed as previously described [68] . pGEX-based plasmids were transformed into E . coli BL21 codon plus ( DE3 ) ( Stratagene ) . pRSET-A- and pET3A-based plasmids were transformed into E . coli BL21 ( DE3 ) /pLysS ( Stratagene ) . GST- and 6xHis-tagged proteins were expressed and purified as previously described [48] . T7-HBZ-bZIP was purified using the T7-Tag Affinity Purification kit ( Millipore-Sigma , 69025 ) according to the manufacturer’s instructions . All purified proteins were dialyzed against 0 . 1M HM ( 50 mM HEPES [pH 7 . 9] , 100 mM KCL , 12 . 5 mM MgCl2 , 1 mM EDTA , 20% [vol/vol] glycerol , 0 . 025% [vol/vol] Tween 20 , 1 mM dithiothreitol [DTT] ) , aliquoted and stored at -80°C . GST pulldown assays were performed as described [58] with the following modifications: glutathione-agarose beads were equilibrated in RIPA buffer containing 1 mM DTT and incubated with 50 pmol of GST-KIX at 4°C for 1 h . Beads were then washed twice with RIPA buffer and combined with whole-cell extract prepared in RIPA buffer . Binding reactions were mixed at 4°C overnight , and beads were subsequently washed four times with RIPA buffer and re-suspended in SDS sample dye . Eluted proteins were analyzed by Western blot . EMSAs using recombinant purified proteins were performed as described with slight modification [48] . Proteins ( combined at amounts stated in the figure legends ) were incubated for one hour at 20°C and then supplemented with 2 to 10 fmol of 32P-end-labeled double-stranded DNA probe , 100 ng of poly ( dA ) ·poly ( dT ) , 1 μg of BSA , and 1 mM DTT in 0 . 5x TM ( 100 mM Tris-HCl [pH 7 . 5] , 20 mM MgSO4 ) . Reactions were then incubated for one hour at 20°C . Protein/DNA complexes were resolved on 5% non-denaturing polyacrylamide gels at 200 volts at room temperature . Probe sequences are indicated in the figures . Transfection of cells , preparation of nuclear extracts and immobilized DNA-binding assays were performed as described with the following three modifications [68] . First , HEK293T cells ( 8 × 106 cells ) were transfected with 25 μg pcDNA-HBZ-Myc-His and/or 25 μg pCMV-MafG-FLAG ( brought to 50 μg of plasmid with the empty vector ) . Second , for each reaction , 7 pmol of biotinylated double-stranded DNA oligonucleotide was bound to the streptavidin beads ( the DNA sequence is shown in the figure ) . Third , bead-bound protein complexes were washed twice with ITB ( 20 mM HEPES [pH 7 . 9] , 0 . 2 mM EDTA , 100 mM KCl , 6 . 25 mM MgCl2 , 10 mM ZnSO4 , 20% [vol/vol] glycerol , 0 . 01% Triton X-100 , 5% BSA , 0 . 2 mM PMSF , 1 mM benzamidine , 10 μg/mL aprotinin , 10 μg/mL leupeptin , 1 mM DTT ) , three times with ITB containing 600 mM KCl , and once with PBS . ChIP assays were performed using the Zymo Spin ChIP Kit ( Zymo Research ) according to the manufacturer’s instructions , but with the following modifications: chromatin was prepared from ~1 x 107 cells , and 150 μg to 200 μg of crosslinked , sonicated chromatin was immunoprecipitated with 5 μg of antibody or normal rabbit serum ( negative control , Jackson ImmunoResearch laboratories , Inc . ) overnight at 4°C . Crosslinked chromatin was sonicated using a Misonix Sonicator 4000 ( 20 sec on , 30 sec off for 5 to 15 min depending on the cell line , amplitude 60 ) . Purified ChIP DNA was analyzed by real-time PCR as described [104] . Primer sequences are as follows: HMOX1 Distal , 5’-CCCTGCTGAGTAATCCTTTCC -3’ and 5’-CTGAGTCACGGTCTAGAGATTTG-3’; HMOX1 Proximal , 5’-CATTTCTGCTGCGTCATGTTT-3’ and 5’-GTAGGCAGGAGGAAGTGAAAC-3’; HMOX1 Gene , 5’-CGCCTTCATGATGAGCATAAC-3’ and 5’-GTTATGCTGTACCTCCTCCTC-3’ . Standard curves were generated for primer sets using 5-fold serial dilutions of each input DNA from the ChIP procedure and were included on each experimental plate . PCR efficiencies ranged from 95–156% , with correlation coefficients >0 . 90 . Enrichment values were quantified relative to the input as described [105] . Results were derived from using the Gene site in pcDNA cells as a normalization factor , which involved comparing ChIP samples that used the same antibody . Specifically , for a given ChIP sample , results were calculated by dividing the enrichment values obtained for that ChIP sample by the enrichment value for the Gene site in pcDNA cells , which set “pcDNA Gene” to 1 . TurboFect ( Thermo Fisher Scientific ) was used to transfect 4 x 106 Jurkat cells with 100 ng of pGL4 . 26 or pGL-4xT-MARE-Luc and 250 ng of the expression vector ( s ) indicated in the figure legend ( the total plasmid quantity was brought to 1 μg with the empty expression vector ) . For results shown in Fig 7B , cells were co-transfected with 100 , 250 and 500 ng of pSG-HBZ-Myc . For results shown in Fig 7F , cells were co-transfected with 162 . 5 , 325 and 650 ng of pcDNA-Nrf2-DN-Myc-His . Cells were processed 24 hours post-transfection using the Luciferase Assay System ( Promega ) according to the manufacturer’s instructions , and luminescence was measured using a Glomax 20/20 Luminometer . For each expression vector/pair of expression vectors , luminescence from cells co-transfected with pGL4 . 26 ( background ) was subtracted from luminescence from cells co-transfected with pGL-4xT-MARE-Luc . Samples were processed as described with minor variations [106] . Cells were collected , washed with PBS , suspended in 50–100 μL of extraction buffer ( 2% TCA , 1 mM EDTA ) , ice-chilled for 15 minutes , vortexed for 45 seconds , and ice-chilled for an additional 15 minutes . Protein concentrations of the cell extracts were determined by Bradford protein assay ( Bio-Rad ) and equalized to 2 mg/mL by diluting samples with extraction buffer . Samples were centrifuged at 4000 x g at 4°C for 10 minutes , and supernatants were analyzed by LC-MS ( the same instruments as above ) . Standards of reduced L-glutathione ( GSH ) ( Millipore-Sigma , G4251 ) and oxidized L-glutathione ( GSSG ) ( Millipore-Sigma , G4376 ) reconstituted in 50 μL of extraction buffer were used for each experiment . Separation was achieved using an Agilent Zorbax Eclipse Plus C18 column 3 . 5 μm , 2 . 1 × 150 mm ) held at 35°C . Mobile phase A consisted of water with 1% formic acid whereas mobile phase B consisted of acetonitrile with 1% formic acid . Flow rate was held at 0 . 1 mL/min throughout analysis . Initial mobile phase composition began at 20% B which was ramped up to 100% B over the next 3 minutes and held constant at 100%B for the following 4 minutes resulting in a total analysis time of 7 minutes . The TOF was operated in positive mode . Reduced and oxidized glutathione were identified and quantified based on the [M+H] and [M+2H] ions respectively . Sample concentrations of GSH and GSSG were determined based on the GSH and GSSG standards . HeLa clones were plated in 96 well plates at 2 . 5 x 104 per well and allowed to adhere overnight . Culture media were then replaced with low serum media ( 0 . 5% serum ) containing 40 μM hemin chloride ( Millipore-Sigma 3741 ) or dimethyl sulfoxide ( DMSO ) ( Millipore-Sigma D2650 ) and cultured for 24 hours . Cell viability was then evaluated using the MTT Cell Growth Assay kit ( Millipore-Sigma ) according to the manufacturer’s instructions . For T-cell lines , viability was assessed using alamarBlue ( Bio-Rad , BUF012 ) according to the manufacturer’s instructions . Cells were plated in 96 well dark plates at 2 . 5 x 104 cells per well in low-serum medium prior to treatment with the indicated concentrations of hemin or DMSO and cultured for 24 hours . Chemical inhibition of HMOX-1 activity was achieved using 1-[[2-[2- ( 4-Bromophenyl ) ethyl]-1 , 3-dioxolan-2-yl]methyl]-1H-imidazole hydrochloride ( OB-24 , Tocris ) . Cells were simultaneously treated with 10 μM OB-24 , or DMSO , and 20 μM hemin chloride , or DMSO , for 24 hours . Fluorescence was detected using a fluorescent microplate reader ( FL600 , Bio-Tek ) . Data are presented in function of vehicle-treated cells for which viability was set to 100% . Microarray data sets used in this study are available at NCBI Gene Expression Omnibus ( GEO ) : accession numbers GSE94409 [59] and GSE55851 [60 , 61] . For the latter , samples were grouped based on patient disease classification and surface expression of CADM1 and CD7 as described [62 , 63] . For each sample , probes corresponding to the HMOX1 transcript were identified and GEO2R was used to obtain expression values . ChIP-Seq data sets GSE31477 [45] were analyzed using UCSC Genome Browser [107 , 108] . Data were visualized using the Human Feb . 2009 ( GRCh37/hg19 ) assembly and samples included GSM935290 ( Stanford_ChipSeq_HeLa-S3_MafK_ ( ab50322 ) _IgG-rab ) . Data from GEO accession number GSE94732 [59] were visualized using the Human Mar . 2006 ( NCBI36/hg18 ) assembly . Alignments between these assemblies were performed manually using NCBI Genome Data Viewer and NCBI BLAST .
|
Human T-cell Leukemia Virus type I ( HTLV-1 ) infection causes a fatal form of leukemia known as Adult T-cell Leukemia ( ATL ) . Given that most current anti-cancer drugs fail to eradicate the leukemic cells in ATL patients , novel treatment strategies are required . One approach is to target functions of one of the proteins produced by the virus , known as HBZ , which is believed to contribute to the development of ATL and to the survival of leukemic cells in ATL patients . In this study , we found that HBZ increases the expression of the cellular enzyme , HMOX-1 , which plays a central role in curbing oxidative stress . In the cell , oxidative stress leads to DNA , protein and lipid damage that can ultimately trigger cell-death . HBZ activates HMOX-1 expression at the level of transcription by commandeering a set of cellular transcription factors known as the small Mafs . We provide evidence that increased HMOX-1 levels in ATL cells may counteract effects of anti-cancer drugs that act by inducing oxidative stress . Additionally , we show that high HMOX-1 levels may dissipate oxidative stress caused by other HTLV-1 proteins , thereby promoting the general survival of HTLV-1-infected cells . Given these results , targeting HMOX-1 might be considered as an ATL patient-treatment option .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"heme",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"hela",
"cells",
"oxidative",
"stress",
"pathogens",
"immunology",
"biological",
"cultures",
"microbiology",
"retroviruses",
"viruses",
"rna",
"viruses",
"cell",
"cultures",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"htlv-1",
"gene",
"expression",
"t",
"cells",
"microbial",
"pathogens",
"cell",
"lines",
"biochemistry",
"cell",
"biology",
"post-translational",
"modification",
"basic",
"leucine",
"zipper",
"domains",
"viral",
"pathogens",
"genetics",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cultured",
"tumor",
"cells",
"antioxidants",
"organisms"
] |
2019
|
HTLV-1 basic leucine zipper factor protects cells from oxidative stress by upregulating expression of Heme Oxygenase I
|
Heterotrimeric G-proteins are molecular switches integral to a panoply of different physiological responses that many organisms make to environmental cues . The switch from inactive to active Gαβγ heterotrimer relies on nucleotide cycling by the Gα subunit: exchange of GTP for GDP activates Gα , whereas its intrinsic enzymatic activity catalyzes GTP hydrolysis to GDP and inorganic phosphate , thereby reverting Gα to its inactive state . In several genetic studies of filamentous fungi , such as the rice blast fungus Magnaporthe oryzae , a G42R mutation in the phosphate-binding loop of Gα subunits is assumed to be GTPase-deficient and thus constitutively active . Here , we demonstrate that Gα ( G42R ) mutants are not GTPase deficient , but rather incapable of achieving the activated conformation . Two crystal structure models suggest that Arg-42 prevents a typical switch region conformational change upon Gαi1 ( G42R ) binding to GDP·AlF4− or GTP , but rotameric flexibility at this locus allows for unperturbed GTP hydrolysis . Gα ( G42R ) mutants do not engage the active state-selective peptide KB-1753 nor RGS domains with high affinity , but instead favor interaction with Gβγ and GoLoco motifs in any nucleotide state . The corresponding Gαq ( G48R ) mutant is not constitutively active in cells and responds poorly to aluminum tetrafluoride activation . Comparative analyses of M . oryzae strains harboring either G42R or GTPase-deficient Q/L mutations in the Gα subunits MagA or MagB illustrate functional differences in environmental cue processing and intracellular signaling outcomes between these two Gα mutants , thus demonstrating the in vivo functional divergence of G42R and activating G-protein mutants .
G protein-coupled receptors ( GPCRs ) convert extracellular signals to intracellular responses , primarily by stimulating guanine nucleotide exchange on heterotrimeric G-protein Gα subunits [1] . Upon receptor-stimulated exchange of GTP for GDP , Gα subunits undergo a conformational change , dominated by three mobile switch regions , resulting in separation of Gα from the obligate Gβγ heterodimer [2] . Switches one and two directly contact the bound guanine nucleotide and include residues critical for catalyzing GTP hydrolysis , while switch three contacts switch two in the activated conformation [3] . The nucleotide-dependent conformational shift of Gα subunits can be monitored biochemically by differential resistance to proteolysis by trypsin or altered tryptophan fluorescence spectra [4] , [5] . The switch mechanism of activation is highly conserved among the mammalian Gα subunit family members , as well as in those found in fungi [6] , [7] . The activated Gα and free Gβγ subunits propagate signals through numerous effectors , including adenylyl cyclases , phospholipases , ion channels , and phosphodiesterases [8] . Mammals express multiple Gα subunits which can be classified into subfamilies according to function . For example , members the Gαi/o subfamily have inhibitory effects on adenylyl cylase and stimulatory effects on cGMP-phosphodiesterase , while Gαq subfamily members stimulate phospholipase C isoforms , promoting hydrolysis of phosphatidylinositol bisphosphate to produce diacylglycerol and inositol triphosphate [9] , [10] . Gα signaling is terminated by intrinsic hydrolysis of bound GTP to GDP , a reaction accelerated by ‘regulators of G-protein signaling’ ( RGS proteins ) , and reversion of the Gα switch conformation to the inactive , GDP-bound state [9] , [11] . Gα·GDP can then re-assemble a heterotrimer with Gβγ or , in the case of the Gαi/o subfamily , engage GoLoco motif proteins that are also selective for the inactive Gα state [12] . In addition to naturally occurring conformationally selective binding partners , phage display peptides have also been engineered to discriminate between Gα·GDP and Gα·GTP . For example , the peptides KB-752 and KB-1753 selectively interact with the inactive GDP-bound and active GTP-bound states of Gαi1 , respectively [13] . Heterotrimeric G-protein signaling components are well-characterized regulators of mammalian biology and are also utilized as sensors for extracellular cues in non-mammalian organisms , such as fungi , plants , and yeast [7] , [14] , [15] . The rice blast fungus , Magnaporthe oryzae , forms infection structures known as appressoria in response to specific environmental surface signals [16] . For example , hydrophobic , but not hydrophilic surfaces , promote appressorium formation [17]–[19] . Genetic studies have implicated a number of G-protein signaling pathway components in the regulation of M . oryzae pathogenesis , including a Gβ subunit ( MGB1 ) [20] , adenylyl cyclase ( Mac1 or MAC ) [21] , cAMP phosphodiesterase ( PdeH ) [22] , and cAMP-dependent protein kinase A ( cPKA ) [23] . M . oryzae also possesses three Gα subunits ( MagA , MagB , and MagC ) with sequence similarity to the Gαs , Gαi , and the fungal-specific GαII subfamilies , respectively [19] , [24] , [25] . Previous studies on Gα subunit deletion strains and magB mutants suggest a role for heterotrimeric G-protein signaling in M . oryzae growth , sexual reproduction , and appressorium formation [24] , [26] . Additionally , an RGS protein ( Rgs1 ) negatively modulates all three M . oryzae Gα subunits [19] . Among the most stringently conserved motifs of Gα subunits is the phosphate-binding loop ( P-loop ) ( Figure S1 ) . Very little variation in the P-loop sequence is seen across Gα subunits in distantly related species , including plants , fungi , and metazoans [27] . In fact , the P-loop is also conserved as a key phosphoryl group-interacting motif in ATP-binding kinases and members of the Ras GTPase superfamily [28] . A P-loop mutation to human Ras isoforms , Gly-12 to valine , is frequently found in human cancers . Ras G12V mutants are GTPase deficient , and thus constitutively active , leading to aberrant signaling and oncogenesis [29] . In fact , mutation of H-Ras Gly-12 to any residue other than proline results in constitutive activity [30] . Mutation of the corresponding P-loop residue in Gαi1 , Gly-42 to valine , also drastically reduces its GTPase activity [31] . Structural studies of Gαi1 ( G42V ) suggest that the introduced valine side chain sterically prevents appropriate positioning of Gln-204 , a residue that coordinates a nucleophilic water molecule during GTP hydrolysis [31] . This glutamine is highly conserved and critical for GTPase activity; its mutation to leucine ( “Q/L” ) in Ras GTPases or Gα subunits also leads to constitutive activity [11] , [29] . Genetic studies of heterotrimeric G-protein function in fungal species have used GTPase deficient Gα Q204L mutants ( referred to as Q/L mutants ) . Additionally , a Gα subunit P-loop mutation , G42R , has been utilized in a similar context . Given that Gαi1 ( G42V ) is GTPase-deficient and mutation of the corresponding glycine in Ras to any amino acid other than proline results in constitutive activation , it has been assumed that G42R mutants would be dominant and constitutively active [32] . Although the biochemical mechanism of the Gα G42R mutant has not previously been characterized , we and others have utilized it to probe the G-protein mediated biology of many fungal species ( Table S1 ) [19] , [26] , [32]–[41] . The phosphate-binding P-loop and switch mechanism of activation are both stringently conserved among Gα subunits from mammals to fungi [6] , [7] ( Figure S1 ) . For example , human RGS2 recognizes the highly similar GTP hydrolysis transition state conformations of both human Gαq and a yeast Gα subunit ( GPA1 ) , such that RGS2 expression complements the deletion of an RGS protein gene in S . cerevisiae [42] , [43] . Furthermore , chimeras of GPA1 and human Gα subunits can function in the yeast pheromone signaling pathway [44] . The residue position corresponding to Gly-42 in Gαi1 is within potential contact distance of residues in the switch regions of the structurally conserved Gα subfamily members [3] , [10] , [45]–[47] . The switch region sequences are highly conserved across mammalian Gα subfamilies , as well as in other species , including M . oryzae , A . nidulans , and S . cerevisiae ( Figure S1 ) . Given the sequence and structural conservation of these regions in Gα subunits , as well as the demonstrated consistent behavior of other point mutations in these regions across multiple Gα subunits ( e . g . the GTPase-deficient Gαi1 ( Q204L ) and the RGS-insensitive Gαi1 ( G184S ) [48] ) , the behavior of the G42R mutation is expected to be consistent in MagA , MagB , and the mammalian Gα subunits . Since we were unable to obtain properly folded recombinant MagA or MagB proteins and no direct cellular assays of MagA or MagB activity are currently available , we utilized three mammalian Gα subunits to investigate the behavior of G42R mutants . Here , we determine through structural , biochemical , genetic , and cellular approaches that Gα subunit G42R mutants are neither GTPase deficient nor constitutively active . Rather , the mutant arginine side chain prevents transition to the activated state upon Gα binding to GTP . Direct phenotypic analyses of M . oryzae strains harboring either Gα G42R mutants or the GTPase-deficient Gα Q204L suggests that a re-evaluation of previous fungal genetic data generated with the G42R mutation is required .
To understand how the G42R P-loop substitution affects Gα subunit structure and function , we obtained a 3 . 0 Å resolution crystal structure model of Gαi1 ( G42R ) bound to GDP using the inactive state-selective phage display peptide KB-752 as a crystallography tool [49] . The asymmetric unit contained three Gαi1 ( G42R ) subunits bound to GDP and Mg2+; two of three monomers were bound to the KB-752 peptide , while the third ( chain C ) lacked electron density for the peptide and instead displayed switch region disorder characteristic of free , GDP-bound Gα subunits [31] . For data collection and refinement statistics , see Table S1 . A comparison of our model with that of wild type Gαi1·GDP/KB-752 ( PDB id 1Y3A ) revealed minor perturbations to the inactive state upon introduction of Arg-42 ( Figure 1A ) . The side chain of Arg-42 projects away from the nucleotide-binding pocket , making no direct contacts with other Gαi1 ( G42R ) residues . Switch 1 and the adjacent β2 strand adopt slightly different conformations in the mutant Gαi1 ( Cα atoms r . m . s . d . 1 . 3 Å ) , likely because the basic residues Arg-178 and Lys-180 are electrostatically and sterically repelled from their wild type orientations by the positively charged Arg-42 side chain ( Figure 1B ) . Arg-178 is known to stabilize the leaving phosphate group during GTP hydrolysis [11]; its perturbation in the Gαi1 ( G42R ) structure model is consistent with the previously assumed GTPase deficiency of G42R mutants . Substitution of the corresponding Gly-12 in H-Ras for any amino acid other than proline yields GTPase deficiency and constitutive activity [30] . Thus it was previously reasoned that Gα ( G42R ) mutants were also incapable of GTP hydrolysis [26] . Binding of GTP by purified Gα subunits can be assessed with the non-hydrolyzable GTP analog , the radionucleotide GTPγ[35S] . Similarly , GTPase activity can be quantified by tracking release of radioactive inorganic phosphate from [γ-32P]GTP-loaded Gα subunits during a single round of hydrolysis [15] . GTPγ[35S] radionucleotide binding and [γ-32P]GTP single turnover hydrolysis assays indicated that the kinetics of GTP binding and hydrolysis by the equivalent G42R mutant GαoA ( G42R ) , in the most frequent splice variant of the mammalian adenylyl cyclase inhibitory Gαo1 , are not significantly different from wild type GαoA ( Figure 2A , B ) . Since the nucleotide binding and hydrolysis rate of this G42R mutant was unexpectedly not perturbed , we further examined the effect of the G42R mutation on Gα interactions with known protein binding partners . RGS proteins accelerate the intrinsic GTPase activity of Gα subunits by stabilizing the transition state for GTP hydrolysis , a conformation mimicked by Gα binding to GDP , AlF4− , and Mg2+ [11] . Surface plasmon resonance ( SPR ) was utilized to detect optical changes upon injection of wild type or G42R mutant GαoA over a surface coated with immobilized GST-RGS12 in the presence of either GDP , GTP , the non-hydrolyzable GTP analog GTPγS , or the hydrolysis transition state-mimetic GDP·AlF4− [50] . The RGS domain of RGS12 bound selectively to wild type GαoA in its GDP·AlF4−-bound state ( KD = 1 . 27±0 . 06 µM ) , as measured by surface plasmon resonance ( SPR ) [50] . However , GαoA ( G42R ) did not engage the RGS domain in any nucleotide state at concentrations up to 25 µM ( Figure 2C , D ) , suggesting that G42R mutants do not adopt a typical GTP hydrolysis transition state in the presence of AlF4− and Mg2+ ( AMF ) , or alternatively that Arg-42 directly interferes with RGS domain binding . A superimposition of Gαi1 ( G42R ) /KB-752 and the Gαi1/RGS4 complex ( PDB 1AGR; not shown ) indicated that the mutant arginine side chain likely directly perturbs the RGS-binding surface . To further characterize nucleotide state-dependent interactions of Gα ( G42R ) , we measured binding affinity toward three additional state-selective Gα-binding partners: Gβγ subunits , a GoLoco motif , and a phage display peptide , KB-1753 [13] . Gα subunits in their GDP-bound , inactive conformations form heterotrimers with Gβγ subunits [6] , and the interaction is disrupted by AlF4− or GTP binding to the Gα subunit . As expected , wild type Gαi1·GDP bound Gβ1γ1 as measured by SPR , but activation of the Gα subunit with GDP·AlF4− prevented association with Gβγ ( Figure 3A ) . However , Gαi1 ( G42R ) engaged Gβ1γ1 in both nucleotide states . Interaction of Gα subunits with fluorophore-labeled peptides was assessed by detecting differences in fluorescence polarization between low molecular weight free peptide and the higher molecular weight Gα/peptide complex [51] . Similar to Gβγ , the GoLoco motif of RGS14 was highly selective for binding the GDP-bound , inactive state of wild type Gαi1 ( KD = 9 . 0±1 . 1 nM ) over the activated GDP·AlF4−-bound form , as determined by fluorescence polarization ( Figure 3B ) . Gαi1 ( G42R ) displayed a much reduced selectivity for RGS14 GoLoco motif binding between the GDP and AlF4− nucleotide states , being only 3-fold selective for the GDP form , whereas wild type Gαi1 is >1000-fold selective . Finally , we tested two G42R mutant nucleotide states for interaction with the active conformation-selective phage display peptide KB-1753 using fluorescence polarization [13] . As expected , KB-1753 selectively interacted with wild type Gαi1·GDP·AlF4− ( KD = 470±40 nM ) relative to GDP-bound Gαi1 ( Figure 3C ) . In contrast , Gαi1 ( G42R ) displayed only weak affinity for KB-1753 in either nucleotide state , as measured by fluorescence polarization . Together these data indicate that Gα ( G42R ) mutants preferentially engage inactive conformation-selective binding partners regardless of the bound nucleotide . To assess the conformational shift of Gα ( G42R ) mutants upon activation with AlF4− or a non-hydrolyzable GTP analog , we utilized intrinsic tryptophan fluorescence and limited trypsin proteolysis . Upon binding GDP·AlF4− or GTP analogs , Gα subunits undergo conformational changes dominated by the three switch regions [52] . A tryptophan residue ( Trp-211 in Gαi1 ) within switch 2 is shifted from a solvent-exposed to a buried orientation , resulting in a reduced efficiency of tryptophan fluorescence quenching that can be detected upon excitation of the Gα protein with light at 284 nm wavelength [5] . Wild-type Gαi1 displayed a large increase in tryptophan fluorescence upon exposure to AlF4− , indicative of a shift to the activated conformation . In contrast , the shift in tryptophan fluorescence of Gαi1 ( G42R ) at the same concentration was blunted relative to wild type and occurred with faster kinetics ( kobs = 0 . 19±0 . 01 s−1 [95% C . I . ] , compared to kobs = 0 . 05±0 . 01 s−1 for wild type Gαi1; Figure 4A ) . The active and inactive states of Gα subunits are also differentially sensitive to proteolysis by trypsin; the more flexible loop conformations of Gα·GDP promote cleavage [4] . While the flexible N-terminus of wild type Gαi1 was cleaved in all three nucleotide states , the resulting ∼38 kDa fragment was resistant to limited trypsin proteolysis in the GDP·AlF4− or GTP-bound conformations relative to the inactive , GDP-bound form ( Figure 4B ) . Gαi1 ( G42R ) , however , was readily proteolyzed in any nucleotide state . Addition of AlF4− had no detectable effect on Gαi1 ( G42R ) resistance to trypsin proteolysis , while GTPγS provided only mild protection of the ∼38 kDa species compared to that of wild type Gαi1 . These data further support the hypothesis that the switch regions of Gα ( G42R ) mutants do not assume appropriate transition state-mimetic or activated state conformations in the presence of AlF4− and GTPγS , respectively . We next sought a structural explanation for the disrupted conformational switch of Gα ( G42R ) mutants . As previously mentioned , the Arg-42 side chain conformation , as modeled in the free GDP-bound Gαi1 ( G42R ) , would not allow glutamine-204 to assume its critical position for orienting the nucleophilic water required for GTP hydrolysis ( Figure 1 ) . However , unlike the G42V mutant of Gα subunits , the G42R mutant retains normal GTP hydrolysis kinetics ( Figure 2 ) . Positioning of Gln-204 for hydrolysis may be possible if the Arg-42 side chain adopts an alternate rotamer . We also crystallized Gαi1 ( G42R ) ·GDP in complex with the GoLoco motif from RGS14 and derived an independent structural model at 2 . 8 Å resolution ( Table S2 ) . In one of the two monomers of the asymmetric unit , Arg-42 adopts such an alternative rotamer that would allow Gln-204 to orient the nucleophilic water for hydrolysis ( Figures 4C and S2 ) . Since we are presently unable to crystallize Gαi1 ( G42R ) in either its GDP·AlF4− or GTP analog-bound states , we superimposed our structural model of Gαi1 ( G42R ) ·GDP ( excluding the RGS14 GoLoco peptide ) with the previously described , wild type Gαi1·GTPγS ( PDB id 1GIA ) ( Figure 4C , D ) . In the activated , GTPγS-bound state of wild type Gαi1 , switches 1 and 2 converge on the nucleotide γ-phosphoryl group , while Glu-236 of switch 3 forms a new polar contact with the backbone of switch 2 [3] . The result is a convergence of the three switch regions near the P-loop to form a stable interface recognized by effector molecules . Superposition of Gαi1 ( G42R ) ·GDP suggests that the bulky Arg-42 side chain would not be easily accommodated by the active switch conformations observed in wild type Gαi1·GTPγS ( Figure 4C , D ) . The arginine as modeled would sterically prevent the positioning of switch 3 residues Leu-234 and Glu-236 as seen in the wild type , activated state . Thus , the Arg side chain likely sterically prevents a normal activated conformation of the switch regions . These data suggest that Arg-42 hinders attainment of the activated switch conformations seen in wild-type Gα subunits , but rotameric flexibility of the mutant side chain allows critical switch residues to effect GTP hydrolysis . Although the G42R mutants of Gα subunits have been shown to favor the inactive conformation despite retaining the ability to bind and hydrolyze GTP , we also sought to investigate their behavior in a cellular context . To investigate the effects of G42R mutants in a signaling pathway context , we introduced the corresponding P-loop mutation into the phospholipase C stimulating mammalian Gα subunit , Gαq ( G48R ) . Wild-type Gαq·GTP activates phospholipase Cβ ( PLCβ ) , which in turn hydrolyzes phosphatidylinositol-4 , 5-bisphosphate ( PIP2 ) to yield diacyl glycerol ( DAG ) and inositol triphosphate ( IP3 ) [10] . Phospholipase C activity can be quantified by measuring accumulation of radioactive IP3 in cells pre-treated with tritiated inositol . Overexpression of wild type Gαq in COS-7 cells had little effect on inositol phosphate accumulation , while the GTPase-deficient and constitutively active Gαq ( Q209L ) markedly stimulated PLCβ activity in a dose-dependent fashion ( Figure 5A , B ) . Gαq ( G48R ) , however , had no significant effect on PLCβ activity when overexpressed , confirming its lack of constitutive activity . Activation of PLCβ by endogenous and overexpressed Gαq can be stimulated by exposure to AlF4− , since Gαq·GDP·AlF4− has high affinity for PLCβ [53] . As expected , endogenous Gαq was activated by AlF4− , and the effect was enhanced by overexpression of wild type Gαq . However , overexpressed Gαq ( G48R ) did not respond to AlF4− stimulation to the same extent as wild type Gαq , reflecting its inability to assume a fully-activated conformation ( Figure 5C , D ) . The Gα ( G42R ) mutant utilized in genetic studies of fungal species , such as Aspergillus nidulans and the rice blast fungus Magnaporthe oryzae , was assumed to be GTPase deficient and thus constitutively active [26] , [32] , and has been used extensively to understand the biology of fungal G-protein signaling [19] , [26] , [32]–[41] . Since the biochemical and structural characterization of such G42R mutants ( Figures 1–4 above ) indicate intact GTPase activity and , instead of constitutive activity , an inability to assume the activated conformation , we sought to clarify the behavior of G42R mutations in the Gα subunits of M . oryzae . We directly compared strains of M . oryzae harboring mutations in the Gα subunits MagA or MagB . Since both Gα subunits are known to regulate appressorium formation in response to inductive , hydrophobic surfaces [24] , we assessed appressorium formation by GTPase-deficient Q/L and non-activatable G42R mutant strains on both hydrophobic and hydrophilic surfaces . The magA ( G45R ) mutant formed slightly fewer appressoria on hydrophobic , inductive surfaces than wild-type M . oryzae , but maintained the differential response to surface hydrophobicity ( Figure 6A , B ) . In contrast , approximately 35% of magA ( Q208L ) conidia formed highly pigmented appressoria , albeit aberrant , after 16 hours , regardless of surface hydrophobicity . The magB ( G42R ) mutant strain resembled magA ( Q208L ) , with ∼30% appressorium formation independent of surface hydrophobicity ( Figure 6C , D ) . The magB ( Q204L ) strain , however , formed very few appressoria on either surface . To further characterize differences between magA and magB G42R and Q/L mutant strains of M . oryzae , we compared colony and conidia morphology , as well as conidiation , to the wild type fungus . Both the magA and magB G42R mutants displayed different overall morphology from the corresponding Q/L mutants ( Figure S3 ) . In the case of magA ( G45R ) , morphology was indistinguishable from the wild type . Upon exposure to light , the magA ( G45R ) also produced slightly fewer conidia when compared to the wild-type M . oryzae , but magA ( Q208L ) formed very few heavily pigmented , aberrant conidia ( Figure 6A , inset and S4A ) . Both magB ( G42R ) and magB ( Q204L ) displayed enhanced conidiation relative to wild type , but those of magB ( Q204L ) were of a distinct morphology , with longer and thinner dimensions than either magB ( G42R ) or wild type ( Figure S4B , C ) . These data indicate that fungal Gα G42R mutants exhibit markedly different phenotypes from truly GTPase-deficient Q/L mutants , consistent with aforementioned structural , biochemical , and cellular experiments that indicate an intact GTPase activity , but a marked inability to achieve an activated conformation . We next determined what effect the introduction of the non-activatable G42R mutant Gα subunits has on fungal infection of barley leaves compared to constitutively active Q/L mutants . As expected , barley leaves inoculated with wild type M . oryzae showed the characteristic dose-dependent formation of disease lesions ( Figure 7 ) . The magA ( G45R ) strain showed similar pathogenicity as the wild type , consistent with intact surface-inducible appressorium formation ( Figure 6B ) . magB ( G42R ) displayed a reduced ability to cause disease , although small lesions were observed at the highest inoculations tested . Both magA ( Q208L ) and magB ( Q204L ) showed drastically reduced lesion formation relative to wild type and the corresponding G42R mutants . These data indicate that constitutive activity of either MagA or MagB can suppress the ability of M . oryzae to penetrate and infect the plant tissue . Additionally , we conclude that the ability of MagB to achieve its activated conformation is critical for Magnaporthe pathogenesis .
Mutant Gα subunit strains have provided excellent tools for probing the functions of heterotrimeric G-proteins in many fungal species , including Aspergillus nidulans and Magnaporthe oryzae ( Table S1 ) [19] , [26] , [32]–[41] . Here , we have demonstrated that the P-loop mutant , G42R , is neither GTPase deficient nor constitutively active as assumed in previous studies . Rather , Gα ( G42R ) is unable to undergo a typical conformational change upon binding GTP , reflected by its inability to engage RGS domains or effector-like molecules . Consistent behavior of Gα ( G42R ) mutations was observed in three mammalian Gα subunit family members: Gαi1 , GαoA , and Gαq . This finding , together with high sequence conservation surrounding the mutant residue ( Figure S1 ) and distinct phenotypes of M . oryzae harboring either Gα ( G42R ) or truly GTPase-deficient Q/L mutants strongly support our hypothesis that MagA ( G45R ) and MagB ( G42R ) are structurally and biochemically similar to the corresponding mammalian Gα mutants . Our crystal structure models of Gαi1 ( G42R ) indicates that this perturbed conformational flexibility is likely due to steric hindrance and electrostatic repulsion between the mutant Arg-42 side chain and residues of the switch regions . The preserved GTPase activity of Gα ( G42R ) mutants implies that Gln-204 is still able to orient a nucleophilic water during GTP hydrolysis . The structural model of Gαi1 ( G42R ) ·GDP bound to the GoLoco motif of RGS14 has provided a snapshot of an alternative Arg-42 rotamer that would indeed allow Gln-204 to access the orientation necessary for GTP hydrolysis . However , this rotamer still is expected to perturb the activated conformation of switch 3 . We conclude that rotameric flexibility at Arg-42 allows the G42R mutant to retain GTPase activity while preventing appropriate active state switch conformations . Interestingly , previous work has identified another Gαi1 point mutation , K180P , that uncouples GTP hydrolysis from nucleotide-dependent conformational change [54] . Gαi1 ( K180P ) is capable of hydrolyzing GTP when not in a fully activated conformation , as also seen for Gαi1 ( G42R ) . Despite the retained ability of Gα ( G42R ) mutants to exchange and hydrolyze nucleotide , they favor an inactive state-like conformation , likely forming a less-dissociable heterotrimer with Gβγ in a cellular context , thereby reducing Gβγ/effector interactions . Since Gα ( G42R ) does not engage effectors with high affinity , it may be expected to behave as a dominant negative mutation; the Gα ( G42R ) /Gβγ heterotrimer may serve as a substrate for receptor-stimulated exchange but fail to activate downstream signaling pathways . In Magnaporthe oryzae , it was previously unclear why strains with magB deleted or expressing the assumedly constitutively active magBG42R exhibited similar phenotypes regarding conidiation , sexual reproduction , and virulence on plant leaves [26] . The present study resolves this issue by demonstrating that the G42R mutant is not constitutively active , but likely exerts a dominant negative effect . The distinct behaviors of Gα ( G42R ) mutants are highlighted by a direct comparison to the truly GTPase-deficient and constitutively active Q/L mutants . Although the magAG45R and magBG42R mutant strains do not reflect constitutive Gα subunit activity , as previously assumed [26] , [32] , they do provide insight into fungal pathogenic development . A phenotypic deficiency upon expression of a Gα ( G42R ) mutant suggests that specific activation of the Gα of interest and subsequent engagement of its downstream effectors is necessary for a particular function of a cell or organism . For instance , both magB deletion [24] and magBG42R mutant strains display drastically reduced induction of appressoria by hydrophobic surfaces , while magA deletion [24] and magAG45R mutations each have minimal effects . Thus , it is likely that MagB transduces an external surface hydrophobicity signal , presumably through a GPCR . Use of the magBG42R mutant suggests that the conformational change accompanying MagB activation is necessary for the selective development of appressoria on hydrophobic surfaces ( Figure S6 ) . It remains to be determined whether the Gα or Gβγ subunits or both propagate signals required for appressorium formation and disease lesion formation in M . oryzae . Direct evidence of interactions between Magnaporthe heterotrimeric G-protein subunits and effector molecules is currently lacking . However , phenotypic similarities between the Gα subunit mutant and deletion strains [20] , [24] , [26] , Gβ subunit ( MGB1 ) deletion [20] , adenylyl cylase ( Mac1 ) deletion [21] , and cAMP phosphodiesterase ( PdeH ) deletion [22] , suggest that MagA and MagB may modulate cellular cAMP level through mechanisms similar to those of mammalian Gαs and Gαi/o . In conclusion , Gα ( G42R ) mutants are incapable of assuming a typical activated conformation , but their retained ability to hydrolyze GTP indicates an uncoupling of conformational change and enzymatic activity . Since G42R mutants are unable to separate from Gβγ or to activate effectors , they provide tools for dissecting the functions of Gα subunits in cellular contexts . Utilizing both G42R and constitutively active Q/L mutants of two Gα subunits , we postulate a critical role for MagB activation in response to growth on hydrophobic surfaces , leading to appressorium formation in the rice blast fungus , M . oryzae .
Unless otherwise noted , all chemicals were the highest grade available from Sigma or Fisher Scientific . Peptides were synthesized by Fmoc ( N- ( 9-fluorenyl ) methoxycarbonyl ) group protection , purified by HPLC , and confirmed using mass spectrometry by the Tufts University Core Facility ( Medford , MA ) . Peptides used for crystallography and biophysical studies have been previously reported: FITC-RGS14 GoLoco [55] , RGS14 GoLoco [56] , FITC-KB-1753 [13] , and KB-752 [49] . Although we were unable to obtain properly folded , purified M . oryzae Gα subunits , the P-loop and surrounding switch regions are highly conserved from mammals to fungi ( Figures S1 ) . Thus , we utilized the readily available purified Gαi1 and GαoA and corresponding G42R mutants . For biochemical experiments , full-length , hexahistidine-tagged Gαi1 and GαoA , and G42R mutants thereof , were purified from E . coli by NTA affinity and gel filtration chromatography as previously described [57] ( see Figure S5 ) . A GST fusion of the RGS12 RGS domain ( aa 664–885 ) was purified as described [58] . Biotinylated Gβ1γ1 was purified as described [59] . For crystallization , an N-terminally truncated ( ΔN30 ) Gαi1 ( G42R ) was expressed and purified by NTA affinity chromatography; the hexahistidine tag was cleaved by TEV protease , and the Gα subunit further purified by ion exchange ( SourceQ , GE Healthcare ) and gel filtration chromatography . Purified Gαi1 ( G42R ) was loaded with excess GppNHp or GDP for 3 hours at room temperature and concentrated to 15 mg/mL in GppNHp crystallization buffer ( 50 mM HEPES pH 8 . 0 , 10 mM MgCl2 , 10 µM GppNHp , 1 mM EDTA , 5 mM DTT ) or GDP crystallization buffer ( 10 mM Tris pH 7 . 5 , 1 mM MgCl2 , 5% v/v glycerol , 5 mM DTT ) . The complex of Gαi1 ( G42R ) and synthetic KB-752 peptide was obtained by mixing a 1∶1 . 5 molar ratio of protein to peptide in GppNHp crystallization buffer . Despite loading of Gαi1 ( G42R ) and crystallization in the presence of GppNHp , the crystal lattice contained Gαi1 ( G42R ) liganded with GDP and bound to KB-752 . The selectivity of KB-752 for the GDP bound state [49] may account for the apparent absence of GppNHp . Crystals of Gαi1 ( G42R ) ·GDP/KB-752 were obtained by vapor diffusion from hanging drops containing a 1∶1 ( v/v ) ratio of protein/peptide solution to well solution ( 17% ( w/v ) PEG MME 5000 , 200 mM MgCl2 , 100 mM HEPES pH 7 . 0 ) . Hexagonal rod crystals ( ∼300×100×100 µm ) formed in 5 days at 18°C exhibited the symmetry of space group P6122 ( a = b = 106 . 6 , c = 455 . 1 , and α = β = 90° , γ = 120° ) and contained two Gαi1 ( G42R ) ·GDP/KB-752 dimers and one Gαi1 ( G42R ) ·GDP monomer in the asymmetric unit . For data collection at 100K , crystals were serially transferred into well solution supplemented with 30% saturated sucrose in 10% increments for ∼30 s , followed by plunging into liquid nitrogen . A native data set was collected at the SER-CAT 22-ID beamline at the Advanced Photon Source ( Argonne National Laboratory ) . Data were processed using the HKL-2000 program [60] . The crystal structure of the wild type Gαi1/KB-752 heterodimer ( PDB 1Y3A [49] ) , excluding the KB-752 peptide , nucleotide , and waters was used as a search model for molecular replacement using the Phaser program [61] . Refinement was carried out using phenix . refine [62] , consisting of conjugate gradient minimization and refinement of individual atomic displacement and translation-libration-screw parameters , interspersed with manual revisions of the model using the Coot program [63] . For data collection and refinement statistics and a list of residues that could not be located in the electron density , see Table S2 . The complex of Gαi1 ( G42R ) and the RGS14 GoLoco motif peptide was generated by mixing a 1∶1 . 5 molar ratio of protein to peptide in GDP crystallization buffer . Crystals of the complex were obtained by vapor diffusion from hanging drops containing a 1∶1 ratio of protein/peptide solution to well solution ( 1 . 7 M ammonium sulfate , 100 mM sodium acetate pH 5 . 0 , 200 mM MgCl2 , 10% ( w/v ) glycerol ) . Crystals ( ∼200×200×50 µm ) formed in 2–5 days at 18°C and exhibited the symmetry of space group C2221 ( a = 70 . 0 , b = 131 . 0 , c = 203 . 3 , and α = β = γ = 90° ) and contained two Gαi1 ( G42R ) /GoLoco motif heterodimers in the asymmetric unit . Diffraction data were collected and processed , and the model refined as described for Gαi1 ( G42R ) /KB-752 , above . The crystal structure of Gαi1 ( Q147L ) /RGS14 GoLoco motif ( PDB 2OM2 [51] ) , excluding the peptide , nucleotide and waters was used as a molecular replacement search model . All structural images were made with PyMOL ( Schrödinger LLC , Portland , OR ) . The [35S]GTPγS filter-binding assay used to measure rates of spontaneous GDP release from wild type and mutant GαoA was conducted as described previously [64] . Intrinsic GTP hydrolysis rates of GαoA and mutants were assessed by monitoring 32P-labeled inorganic phosphate production during a single round of GTP hydrolysis as described previously [65] . Optical detection of protein/protein interactions by surface plasmon resonance was performed using a Biacore 3000 ( GE Healthcare ) . Carboxymethylated dextran ( CM5 ) sensor chips ( GE Healthcare ) with covalently bound anti-GST antibody surfaces were created as described previously [50] . The GST-RGS12 RGS domain protein and GST alone ( serving as a negative control ) were separately immobilized on SPR chip surfaces . Biotinylated Gβ1γ1 and mNOTCH peptide ( serving as a negative control ) were separately immobilized on a streptavidin ( SA ) sensor chip ( GE Healthcare ) as described previously [50] . All polarization experiments were conducted using a PHERAstar microplate reader ( BMG Labtech , Offenburg , Germany ) , essentially as described previously [51] . Changes in tryptophan fluorescence of Gαi1 subunits were measured to assess activation by GDP·AlF4− , as described previously [51] . Activation of Gα subunits results in translocation of a conserved switch 2 tryptophan into a hydrophobic pocket , increasing the quantum yield of tryptophan fluorescence [5] . Fluorescence intensity traces shown are representative of triplicate experiments . Gα subunits are relatively protected from trypsin-mediated proteolysis in the GDP·AlF4− and GTP analog-bound , activated states [4] . Ten µg of wild type or mutant Gαi1 in 50 mM HEPES ( pH 8 . 0 ) , 1 mM EDTA , 5 mM DTT , 0 . 05% C12E10 , and 10 mM MgCl2 were incubated for three hours at room temperature with either 100 µM GDP , 100 µM GTPγS , or 100 µM GDP , 20 mM NaF , and 60 µM AlCl3 . Five hundred ng of N-Tosyl-L-phenylalanine chloromethyl ketone ( TPCK ) -treated trypsin was added to each reaction , followed by a 10-minute incubation at room temperature . Proteolysis was stopped by addition of SDS-PAGE sample buffer and boiling . Samples were subjected to SDS-PAGE and stained with Coomassie Blue . COS-7 cells in 12-well culture dishes were transfected with KT3-tagged wild type or mutant Gαq , metabolically labeled with 1 µCi of [3H]inositol/well and assayed for inositol phosphate accumulation using Dowex chromatography as described previously [66] . For AlF4− stimulation experiments , final concentrations of 10 mM NaF and 30 µM AlCl3 were added to cell media . To determine wild type and mutant Gαq expression levels , cells were lysed in SDS-PAGE sample buffer . Proteins separated by electrophoresis were immunoblotted with anti-KT3 antibody ( Covance ) or anti-actin antibody ( Sigma ) . The M . oryzae wild-type strain B157 was obtained from the Directorate of Rice Research ( Hyderabad , India ) . Magnaporthe strains carrying individual point mutations in the Gα subunits , namely: magAG45R , magAQ208L , magBG42R , magBQ208L have been described previously together with the rgs1Δ mutant [19] . Wild type and mutant strains cultures were maintained at 28°C in the dark on Prune Agar medium plates ( PA; per L: 40 mL prune juice , 5 g lactose , 5 g Sucrose , 1 g yeast extract and 20 g agar , pH 6 . 5 ) . Assessment of the radial growth , aerial hyphae and colony characteristics was carried out as previously described [22] . Conidiation was induced in the Magnaporthe colonies through exposure to continuous incandescent light at room temperature for 6 days . Conidia were harvested by scraping the surface growth in water with an inoculation loop . The suspension was filtered through two layers of Miracloth ( Calbiochem , San Diego , USA ) , collected in Falcon tubes ( BD Biosciences , USA ) , vortexed for a minute to ensure complete detachment of conidia from the mycelia , and then pelleted by centrifugation at 3 , 000 rpm for 15 minutes . The conidia were washed twice and re-suspended in a fixed volume of sterile water . Prior to harvesting the spores , the radius of each colony was measured to calculate the surface area of the colony . Conidia produced by a given colony were quantified using a hemocytometer and reported as the total number of conidia present per unit area of the colony . Droplets ( 20 µl containing 500 conidia ) of conidial suspension were placed on plastic cover slips ( hydrophobic surface ) or hydrophilic side of GelBond membrane ( Lonza Walkersville Inc . , USA ) and incubated in a humid chamber at room temperature . The total number of appressoria formed by each strain on either surface was quantified at 16 hpi ( hours post inoculation ) . For pathogenicity assays , leaves from two week old barley seedlings were cut into smaller pieces ( 2–3 cm long ) and washed in sterile water , following which the leaf bits were rinsed for 45 seconds in 40% ethanol . The leaf pieces were then washed twice with sterile antibiotic-containing distilled water . The washed leaves were placed on kinetin agar plates ( 2 mg/mL kinetin , 1% agar ) . Conidia were quantified and a dilution series of the conidial suspension was inoculated on detached barley leaves at the required concentrations . The samples were incubated in a humidified growth chamber with a 16 h light/8 h dark cycle at 22°C . Disease symptoms were assessed 5–7 days post inoculation . Samples were observed on a BX51 ( Olympus , Japan ) microscope equipped with UPlan FL N 60X/1 . 25 Oil objective with appropriate filter sets . Bright field images were captured using a Cool SNAP HQ camera ( Photometrics , USA ) and processed using Image J ( National Institutes of Health , USA ) , MetaVue ( Universal Imaging , USA ) and Adobe Photoshop 7 . 0 ( Adobe Inc , USA ) .
|
Heterotrimeric G-proteins function as molecular switches to convey cellular signals . When a G-protein coupled receptor encounters its ligand at the cellular membrane , it catalyzes guanine nucleotide exchange on the Gα subunit , resulting in a shift from an inactive to an active conformation . G-protein signaling pathways are conserved from mammals to plants and fungi , including the rice blast fungus Magnaporthe oryzae . A mutation in the Gα subunit ( G42R ) , previously thought to eliminate its GTPase activity , leading to constitutive activation , has been utilized to investigate roles of heterotrimeric G-protein signaling pathways in multiple species of filamentous fungi . Here , we demonstrate through structural , biochemical , and cellular approaches that G42R mutants are neither GTPase deficient nor constitutively active , but rather are unable to transition to the activated conformation . A direct comparison of M . oryzae fungal strains harboring either G42R or truly constitutively activating mutations in two Gα subunits , MagA and MagB , revealed markedly different phenotypes . Our results suggest that activation of MagB is critical for pathogenic development of M . oryzae in response to hydrophobic surfaces , such as plant leaves . Furthermore , the lack of constitutive activity by Gα ( G42R ) mutants prompts a re-evaluation of its use in previous genetic experiments in multiple fungal species .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"mycology",
"proteins",
"biology",
"microbiology"
] |
2012
|
A P-loop Mutation in Gα Subunits Prevents Transition to the Active State: Implications for G-protein Signaling in Fungal Pathogenesis
|
To evaluate antigen-specific immune responses for leprosy diagnosis in a hyperendemic area in China . Eighty-three leprosy patients and 161 non-leprosy controls were enrolled from Hani-yi Autonomous Prefecture of Honghe , Yunnan Province , China . Leprosy patients were divided into multibacillary ( MB , n = 38 ) , paucibacillary ( PB , n = 23 ) , and post-multi-drug therapy ( MDT , n = 22 ) groups . Controls were divided into the following groups: healthy household contacts ( HHC , n = 119 ) , tuberculosis ( TB , n = 11 ) , and endemic controls ( EC , n = 31 ) . The NDO-LID Rapid Test , M . leprae antigen-specific ELISA and antigen-specific IFN-γ secretion in a whole blood assay ( WBA ) were used to evaluate these subjects . The NDO-LID Rapid Test achieved higher positive response rates in MB than in PB patients[94 . 7% ( 36/38 ) vs 65 . 2% ( 15/23 ) ] , and these rates were higher than those observed by ELISA using anti-LID-1[92 . 1% ( 35/38 ) vs 52 . 2% ( 12/23 ) ] , anti-NDO-LID[92 . 1% ( 35/38 ) vs 47 . 8% ( 11/23 ) ] , and anti-ND-O-BSA[89 . 5% ( 34/38 ) vs 60 . 9% ( 14/23 ) ] . However , the NDO-LID Rapid Test also showed a higher positive response rate in the EC group ( 33 . 3% , 10/31 ) , which was higher than the rates observed for anti-NDO-LID ( 12 . 9% , 4/31 ) and anti-ND-O-BSA ( 16 . 1% , 5/31 ) . M . leprae antigen-specific ELISA demonstrated relatively high specificity ( 86 . 84–97 . 37% ) but low sensitivity ( 15 . 97–72 . 73% ) in discriminating between leprosy patients and non-leprosy controls by ROC curve analysis . In contrast , M . leprae antigen-specific IFN-γ secretion detection achieved higher positive response rates in PB than in MB patients ( positive ratio of MB vs PB: 40% vs 56% for LID-1 , 28 . 6% vs 47 . 8% for ML89 , 31 . 4% vs 60 . 7% for ML2044 , and 31 . 4 vs 47 . 8% for ML2028 ) and could distinguish MB from EC when stimulated with ML89 ( AUC = 0 . 6664 ) and PB fromTB when stimulated with ML2044 and ML2028 ( AUC = 0 . 7549 and 0 . 7372 , respectively ) . The NDO-LID Rapid Test and M . leprae antigen-specific ELISA are useful tools to assist in the diagnosis of leprosy patients , especially MB patients , although the former had higher sensitivity but lower specificity than the latter . M . leprae antigen-specific IFN-γ release assessed by WBA has diagnostic value for distinguishing PB from TB but not for distinguishing PB from HHC or EC . Screening novel M . leprae-specific antigens , combining different M . leprae antigens and a multi-cytokine analyte model may be needed for more effective diagnosis of leprosy .
Leprosy , a chronic disease caused by Mycobacteriumleprae ( M . leprae ) infection , has a wide range of clinical outcomes correlated with the host's immune response to the bacilli[1 , 2] . Current leprosy control strategies rely on diagnosing the disease as early as possible , followed by prompt treatment with multi-drug therapy ( MDT ) [1] . The implementation of World Health Organization ( WHO ) MDT for widespread , worldwide treatment has drastically reduced registered leprosy cases from the approximately 12 million reported in 1985 to fewer than 250 , 000 reported in 2006[3] . Currently , leprosy is mainly diagnosed by clinicians using defined criteria , slit-skin smears and biopsies[4] . However , as the prevalence of the disease decreases , clinical expertise is diminishing , leading to extended delays between the onset of clinical signs and the diagnosis and consequent sustained transmission of M . leprae[5] . Leprosy patients are predominantly diagnosed by the appearance of disease signs , but they can also be characterized by the physical and histological attributes of skin or nerve lesions or by their immune response to crude or recombinant M . leprae antigens[6 , 7 , 8 , 9] . It has been demonstrated that the immune response to crude or recombinant M . leprae antigens is helpful for detecting multibacillary ( MB ) leprosy patients by their antibody response[6] , for the diagnosis of paucibacillary ( PB ) patients by antigen-specific CMI[7] , and for monitoring the effectiveness of MDT in MB and PB leprosy patients by the antibody response and antigen-specific CMI , respectively[8] . The M . leprae antigens used for ELISA in this study were Leprosy IDRI diagnostic-1 ( LID-1 ) , a fusion protein developed by fusing the ML0405 and ML2331 genes[9 , 10];NDO-LID , a conjugate of LID-1 with natural octyl disaccharide ( NDO ) [11];and ND-O-BSA , a synthetic PGL-I derivative . The NDO-LID Rapid Test in lateral flow-based format has been developed using NDO-LID . The single tetravalent 89-kDa fusion protein ( ML89 ) , designated LEP-F1 , consists of the ML2028 , ML2055 and ML2380 antigens[12] . A list of accession numbers/ID numbers for genes and proteins included in the NCBI search and mentioned in the text is shown inTable 1 . The purpose of this study was to evaluate the diagnostic value of three antigen-specific immune diagnostic tests , namely , the NDO-LID Rapid Test ( antibody response ) , an antigen-specific enzyme linked immunosorbent assay ( ELISA ) ( anti-LID-1 , anti-NDO-LID , and anti-ND-O-BSA ) ( antibody response ) , and antigen-specific IFN-γ secretion in a whole blood assay ( WBA ) ( stimulated by LID-1 , ML89 , ML2044 and ML2028 ) ( antigen-specific CMI ) for diagnosing leprosy in a hyperendemic area in China .
This study was approved by the Medical Ethics Committee of Beijing Friendship Hospital , Capital Medical University , Beijing , P . R . China . Written informed consent was obtained from all adult participants , and all parents or guardians of child participants provided informed consent on their behalf . All of the procedures in the study involving human participants were performed in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards . Eighty-three leprosy patients , who were referred to the Honghe Prefecture Skin Disease Prevention and Cure Institute in Honghe Autonomous Prefecture , Yunnan Province , were included in the study . Leprosy diagnosis was established based on clinical signs and symptoms , skin smears , skin biopsy , and neuro-physiologic examinations . The leprosy patients were classified into five groups based on the Ridley and Jopling[13] classification: tuberculoid ( TT ) , borderline-tuberculoid ( BT ) , borderline-borderline ( BB ) , borderline-lepromatous ( BL ) , and lepromatous ( LL ) groups . For data analysis in this study , leprosy patients were also classified into three groups: PB and MB , according to the WHO operational classification[14] during MDT , or post-MDT . One hundred and sixty-one controls from the same endemic region were included as non-leprosy controls . The controls were further classified into three groups: healthy household contacts ( HHC ) , tuberculosis ( TB ) , and endemic controls ( EC ) . Antigen-specific antibody detection by NDO-LID was performed as previously described[15] . Serum antibodies were measured by the NDO-LID rapid diagnostic test ( RDT; procured from Orange Life , Rio de Janeiro , Brazil ) . Briefly , NDO-LID RDT was performed by first adding undiluted serum ( 10 μl ) into the sample well within the test cassette , followed by the addition of running buffer ( 100 μl ) . Samples migrated through the cassette such that interactions with the test and/or control lines were revealed as red colored lines within the reading window . Tests were valid if the control line was observed . A positive result was defined by the presence of the test line . Visual results were interpreted after 20 minutes by two independent readers and scored subjectively as ( ±/+/++/+ + + ) , with faint ( ± ) or no test line considered a negative result . ELISA microplate wells were coated overnight with the M . leprae-specific antigens LID-1 ( 1 μg/ml ) , NDO-LID ( 200 ng/ml ) or synthetic PGL-I ( 200 ng/ml ND-O-BSA ) in 0 . 1 M carbonate/bicarbonate coating buffer , pH 9 . 6 ( 50 μl ) . After 1 h in blocking buffer ( 1% bovine serum albumin in phosphate-buffered saline , pH 7 . 2 , with 0 . 05% tween and 1% BSA/PBS/T ) , sera were diluted in blocking solution , tested at a 1:200 dilution ( 100 μl ) , and subsequently incubated for 2 h at room temperature ( RT ) . Then , the wells were washed with PBS with 0 . 05% tween 20 ( PBS/T , wash buffer ) six times . Secondary peroxidase-conjugated anti-human IgM ( anti-PGL-I ) , anti-human IgG ( anti-LID-1 ) ( 1:20 , 000 , Abcam , Cambridge , UK ) , or a combination of anti-human IgM and IgG antibodies ( anti-NDO-LID ) was added for another 2-h incubation period . Following this incubation , the wells were washed with PBS/T six times , followed by the addition of 100 μl of substrate ( 3 , 3’ , 5 , 5’-tetramethylbenzidine; TMB ) . After 15-minute incubation at RT , 50 μl of stop solution ( H2SO4 , 1 M ) was added . Optical density ( OD ) values were determined with an ELISA plate reader ( Asys Expert Plus-Microplate Reader UK ) at 450 nm . The cut off for ELISA positivity was calculated from an OD value of 0 . 2 , as described previously[15] . WBA was performed as previously described . Briefly , undiluted , heparinized venous whole blood ( Greiner ) was collected . Whole blood was plated into 24-well plates ( 450 μl/well; Sigma , St . Louis , MO ) within 2 h of collection and incubated with stimulants for 24 h at 37°C and 5% CO2 . Each assay included stimulation with individual M . leprae recombinant proteins , including LID-1 , ML89 , ML2044 , and ML2028 ( provided by Dr . M . S . Duthie , Infectious Disease Research Institute ( IDRI ) , Seattle , USA ) , at 100 μg/ml in PBS for experimental evaluations or 750 μg/ml PHA ( Sigma ) as a control treatment . Approximately 150 μl of plasma was collected and stored at -20°C until IFN-γ assessment . IFN-γ concentration was determined by ELISA according to the manufacturer’s instructions ( U-CyTech Biosciences Human IFN-γ ELISA kit , CT201A , The Netherlands , CM ) . The IFN-γ ELISA employed had a detection limit of 2 pg/ml , and a threshold for positive responses was arbitrarily selected at 50pg/ml according to a previous study[15] . Statistical analysis was performed primarily with GraphPad Prism software version 5 . 0 ( GraphPad Software Inc . , San Diego , CA , USA ) . The nonparametric Mann-Whitney U test was used to analyze differences between two groups . The Kruskal-Wallis test was used to analyze differences among three or more groups . Probability ( p ) values less than 0 . 05 were considered significant . The diagnostic utility of individual M . leprae antigen-specific responses for leprosy disease , including sensitivity , specificity , Youden’s index , and area under the receiver operator characteristic curve ( AUC ) , were ascertained by receiver operator characteristics ( ROC ) curve analysis . The concordance between results was determined by kappa values ( κ ) , and p values were calculated ( Statistical Package for the Social Sciences ( SPSS ) version 16 . 0 ) .
The study was undertaken mainly in counties in Honghe Autonomous Prefecture , Yunnan ( YN ) Province , southwest China . Other cases were enrolled from the nearby autonomous prefectures of Chuxiong , Zhaotong and Kunming ( provincial capital city ) in YN . Honghe Autonomous Prefecture hadan estimated population of 4 , 470 , 000 in 2015 and is considered highly endemic for leprosy in China ( annual new case detection rate of 1 . 13/100 , 000 from 2000–2007 ) . According to data from the Honghe Prefecture Skin Disease Prevention and Cure Institute , 190 new cases were reported from 2010 to 2014[16] . Eighty-three leprosy cases[MB , n = 38; PB , n = 23; and MDT , n = 22] and 161 controls [HHC , n = 119; TB , n = 11; and EC , n = 31] from the same endemic region were included . The basic information for each study group is summarized in Table 2 . Serum samples were evaluated by the NDO-LID Rapid Test , M . leprae antigen-specific ELISA , and M . leprae antigen-specific secretion of IFN-γ in WBA based on the positive response rate ( Table 3 ) . For the NDO-LID Rapid Test , the positive response rates were higher in the MB than in the PB group[MB vs PB: 94 . 7% ( 36/38 ) vs 65 . 2% ( 15/23 ) ] . For M . leprae antigen-specific ELISA , a trend similar to that observed for the NDO-LID Rapid Test was noted: the positive response rates were also higher in the MB than in the PB group[MB vs PB: 92 . 1% ( 35/38 ) vs 52 . 2% ( 12/23 ) against LID-1 , 92 . 1% ( 35/38 ) vs 47 . 8% ( 11/23 ) against NDO-LID , and 89 . 5% ( 34/38 ) vs 60 . 9% ( 14/23 ) against ND-O-BSA] . Both methods also demonstrated higher response rates in the MB group than in the post-MDT , HHC , EC , and TB groups . For WBA , however , the positive response rates were higher in the PB group than in the MB group[MB:PB: 40% ( 14/35 ) vs 56 . 5% ( 13/23 ) for LID-1 , 28 . 6% ( 10/35 ) vs 47 . 8% ( 11/23 ) for ML89 , 31 . 4% ( 11/35 ) vs 60 . 7% ( 14/23 ) for ML2044 , and 31 . 4% ( 11/35 ) vs 47 . 8% ( 11/23 ) for ML2028] . WBA also showed higher response rates in the PB group than in the post-MDT , HHC , EC , and TB groups , except for the ML89 antigen in the post-MDT and EC groups . When the same samples were evaluated using the NDO-LID Rapid Test , confirmation was achieved in 94 . 7% ( 36/38 ) of MB patients , and a high degree of agreement was observed between LID-1 ( 92 . 1% ) , NDO-LID ( 92 . 1% ) , and ND-O-BSA ( 89 . 5% ) ELISA . For PB patients , the NDO-LID Rapid Test reached 65 . 2% confirmation , which was slightly higher than the results obtained for LID-1 ( 52 . 2% ) , NDO-LID ( 47 . 8% ) , and ND-O-BSA ( 60 . 9% ) ELISA ( Table 2 ) . However , the NDO-LID Rapid Test showed positive responses in 33 . 3% ( 10/31 ) of the EC group , which was similar to the rate for ND-O-BSA ( 38 . 7% , 12/31 ) but higher than those for LID-1 ( 12 . 9% , 4/31 ) and NDO-LID ( 16 . 1% , 5/31 ) ( Table 3 ) . This finding indicates that the NDO-LID Rapid Test is more sensitive than M . leprae-specific antigen ELISA ( anti-LID-1 and anti-NDO-LID ) for detecting leprosy patients , especially MB patients , but has reduced specificity . A kappa test analyzes for the agreement of results collected from various test formats . When a kappa test was performed between the NDO-LID Rapid Test and M . leprae antigen-specific ELISA and between the NDO-LID Rapid Test and WBA , good agreement was only observed between the NDO-LID Rapid Testand M . leprae antigen-specific ELISA ( anti-LID-1 , anti-NDO-LID , and anti-ND-O-BSA ) , with indexes of 0 . 868 , 0 . 868 and 0 . 842 , respectively ( p values of 0 . 000 , 0 . 000 , and 0 . 000 , respectively ) for the MB group ( Table 4 ) . This finding indicates that the two tests showed high consistency for the diagnosis of MB leprosy patients . For all three M . leprae antigens ( LID-1 , NDO-LID and ND-O-BSA ELISA ) , the OD values showed significant differences for MB vs the PB , post-MDT , HHC , TB or EC groups , and PB vs EC ( Fig 1 ) . Of note , NDO-LID was better than the other two antigens ( LID-1 and ND-O-BSA ) at discriminating PB leprosy patients from non-leprosy controls ( Table 5 ) . In addition , we evaluated the diagnostic ability of M . leprae antigen-specific ELISA using ROC curve analysis , AUC , sensitivity and specificity ( Table 6 ) and demonstrated that this method had a relatively high specificity but low sensitivity . We compared the analyte levels detected in M . leprae antigen-stimulated WBA supernatants in leprosy patients with the levels obtained from the non-leprosy control groups using the mean and standard deviation ( SD ) ( Fig 2 ) and the median and range ( Table 7 ) . As described previously , newly diagnosed PB patients produce more IFN-γ than MB patients . We also evaluated the diagnostic potential of IFN-γ by ROC curve analysis and AUC . IFN-γ levels were significantly different in ( 1 ) MB vs EC when stimulated with ML89 ( AUC = 0 . 6664 ) ; ( 2 ) PB vs TB when stimulated with ML2044 and ML2028 ( AUC = 0 . 7549 and 0 . 7372 , respectively ) ; ( 3 ) post-MDT vs TB when stimulated with LID-1 ( AUC = 0 . 8347 ) ; ( 4 ) HHC vs TB when stimulated with LID-1 ( AUC = 0 . 6834 ) ; and ( 5 ) EC vs TB when stimulated with LID-1 , ML89 , ML2044 and ML2028 ( AUC = 0 . 8211 , 0 . 8152 , 0 . 7830 , and 0 . 7361 , respectively ) ( Fig 3 , Table 8 ) .
Widespread application of MDT therapy has led to major advances in leprosy control , with sharp declines in prevalence rates in the vast majority of countries over the last 20 years[15 , 17] . However , the disease remains a public health concern in many regions . In 2010 , China reported 1324 new cases of leprosy to the WHO[18] . The majority of cases in China came from the ethnically diverse , mountainous , and underdeveloped southwest provinces of Yunnan , Guizhou , and Sichuan[19 , 20] . Honghe Autonomous Prefecture inYunan is considered a highly endemic area for leprosy in China . We enrolled 83 leprosy patients and 161 controls from this endemic region in this study to evaluate the ability of several diagnostic tests to correctly diagnose different categories of leprosy patients . We found that the NDO-LID Rapid Test and M . leprae antigen-specific ELISA was useful to diagnose leprosy patients in hyperendemic areas of leprosy disease , especially MB patients . The former method provides a point-of-care measurement of antibodies and had higher sensitivity but lower specificity than the latter . M . leprae antigen-specific IFN-γ secretion in WBA has diagnostic value for distinguishing PB from TB but not for distinguishing PB and HHC or EC . M . leprae-specific antigen tests have been developed as useful tools to diagnose leprosy . LID-1 , NDO-LID , and ND-O-BSA ( also named PGL-I ) , as representative M . leprae-specific antigens , have been widely evaluated as leprosy diagnostics in the hyper-endemic regions of Brazil[20–23] , Colombia , the Philippines[24] , and China[15] and have been demonstrated to be excellent tools for detecting MB leprosy patients in a simple and highly quantitative manner[24] , predicting patients susceptible to developing leprosy type 2 reactions ( T2R ) [23] , and distinguishing leprosy from other confounding dermatoses[18] . The NDO-LID Rapid Testwas compared with M . leprae antigen-specific ELISA and demonstrated a high degree of sensitivity but significant differences in specificity for leprosy diagnosis[25] . Therefore , this test is an effective tool for screening and identifying individuals at high risk who might benefit from regular monitoring[26] . Our previous study showed that confirmation was achieved in 95% of MB leprosy patients with the NDO-LID Rapid Test , and a high degree of agreement was observed with LID-1 , NDO-LID , and ND-O-BSA ELISA[15] . In addition , 63 . 6% of PB leprosy patients were confirmed , and the NDO-LID Rapid Test had a higher detection rate in PB leprosy patients than LID-1 , ND-O-BSA , and NDO-LID ELISA[15] . In this study , we enlarged the sample size and obtained results similar to those of previous studies . These data indicate an improved capacity of the NDO-LID Rapid Test over M . leprae ELISA for detecting the disease . However , the test also suffers from higher positive responses in the EC group than did NDO-LID and LID-1 ELISA . This implies that the NDO-LID Rapid Test was more sensitive but less specific than M . leprae antigen-specific ELISA ( anti-LID-1 and anti-NDO-LID ) for discriminating the leprosy patient group fromthen on-leprosy EC group . Despite the relatively low specificity , the NDO-LID Rapid Test , as a low-tech , robust assay , can still be applied in resource-poor settings to measure the immune response to M . leprae infection and can be used as a tool for leprosy screening in combination with good specificity confirmation tests , which will lead to timely treatment and reduced transmission[27] . We also evaluated the capacity of M . leprae antigen-specific ELISA to discriminate between the leprosy and control groups . All three M . leprae antigens ( LID-1 , NDO-LID and ND-O-BSA ) were able to discriminate the MB group from all other leprosy and non-leprosy groups and the PB leprosy group from the non-leprosy EC groups , whereas only NDO-LID was able to discriminate the PB leprosy group from the non-leprosy HHC group . This indicates that all three M . leprae antigens have potential and specific value for research and medical applications . As described before , the M . leprae antigen-specific ELISA had lower sensitivity but better specificity than the NDO-LID Rapid Test . ELISA detection of specific antibodies may be preferred for confirming diagnoses , differentiating leprosy from other dermatological conditions , and performing follow-up studies for leprosy HHC and indeterminate leprosy , which are very early signs of the disease that are often missed by family members and medical personnel in the endemic area[27] . Cytokines such as IFN-γ have recently been studied as diagnostic host biomarkers for leprosy . M . leprae-specific antigens , such as M . Leprae crude antigens ( M . leprae cell sonicate , MLCS ) , M . leprae recombinant protein ( rML ) ( LID-1 ) , M . leprae diffusion proteins[46f ( ML0405+ML0568 ) and 73f ( ML2028+ML2346+ML2044 ) ] and combinations of rML ( 46f+LID-1 , ML0276+LID-1 , ML2055+ML1632+ML2044 , ML0276+46f , and ML2055+LID-1 ) were used in these studies[20 , 28–31] . IFN-γ and CXCL10 were evaluated as potential diagnostic host markers for PB leprosy patients in the hyper-endemic regions of Brazil[20 , 28–31] and China[15] . Newly diagnosed PB patients produced more IFN-γ than MB patients[28–31] , and IFN-γ was helpful in the differential diagnosis of leprosy from other confounding dermatoses[2] . CXCL10 discriminated PB from EC only in ML0276+LID-1 WBA; however , CXCL10 could not discriminate active disease ( PB ) from HHC individuals[28] . In this study , we also demonstrated that for new cases , PB patients produced more IFN-γ than MB patients; however , IFN-γ did not discriminate active disease ( PB ) from HHC or EC individuals . In this study , we investigated the accuracy of IFN-γ as a host marker detected in supernatants after stimulation of whole blood with M . leprae-specific antigens ( LID-1 , ML89 , ML2044 , and ML2028 ) in an overnight culture assay and compared the IFN-γ marker levels in leprosy and non-leprosy control groups . Although IFN-γ can be useful as a host biomarker that contributes to a diagnostic signature of MB vs EC and that distinguishes PB vs TB groups , there was no evidence that IFN-γ was able to discriminate between the PB and HHC or EC groups . To screen novel M . leprae-specific antigens , combining different M . leprae antigens and facilitating a multi-cytokine analyte model may achieve improved diagnostic potential . This study is limited by the small sample size , especially in the PB group . Antigen-specific immune responses have had limited diagnostic ability for leprosy disease and until recently have only been used for seroepidemiological investigation in hyperendemic areas of leprosy disease or in patients clinically suspected of having leprosy disease . However , the results should be interpreted with caution . Only a very limited number of M . leprae-specific antigens ( LID-1 , ML89 , ML2044 , and ML2028 ) and only one potential diagnostic host biomarker ( IFN-γ ) were tested for leprosy diagnosis in this study . Future studies should focus on additional M . leprae-specific antigens as well as additional host biomarkers . In conclusion , the NDO-LID Rapid Test and M . leprae antigen-specific ELISA were helpful for diagnosing leprosy in hyperendemic areas of leprosy disease , especially for MB patients . The former had higher sensitivity but lower specificity than the latter . Although IFN-γ has been widely studied as a potential biomarker for PB leprosy patients , more research is needed to identify feasible M . leprae-specific antigens and other appropriate host biomarkers to improve its diagnostic value in PB patients in future studies .
|
Although the implementation of World Health Organization ( WHO ) multidrug therapy ( MDT ) treatment has drastically reduced the number of registered leprosy cases , new case detection rates have stabilized over the last decade , and leprosy remains an important health problem in many regions . Antigen-specific immune diagnostic tools are helpful for leprosy diagnosis but require broad evaluation in different populations from areas with hyperendemic leprosy . The NDO-LID Rapid Test , M . leprae antigen-specific ELISA and antigen-specific secretion of IFN-γ in a whole blood assay ( WBA ) can be used to diagnose multibacillary ( MB ) and paucibacillary ( PB ) leprosy patients . The authors found that in Honghe Autonomous Prefecture , Yunnan Province , China , the NDO-LID Rapid Test and M . leprae antigen-specific ELISA have the potential to be used as tools to assist in the diagnosis of patients with MB leprosy . The NDO-LID Rapid Test has higher sensitivity but lower specificity than the M . leprae antigen-specific ELISA . M . leprae antigen-specific IFN-γ secretion in WBA exhibited diagnostic value for distinguishing PB from TB but not for distinguishing PB from HHC or EC . This study provides an evaluation of antigen-specific immune responses for leprosy diagnosis in a hyperendemic area in China .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"mycobacterium",
"leprae",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"immunology",
"tropical",
"diseases",
"biomarkers",
"bacterial",
"diseases",
"neglected",
"tropical",
"diseases",
"immunologic",
"techniques",
"bacteria",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"tuberculosis",
"immunoassays",
"actinobacteria",
"immune",
"response",
"biochemistry",
"diagnostic",
"medicine",
"tuberculosis",
"diagnosis",
"and",
"management",
"leprosy",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2018
|
Evaluation of antigen-specific immune responses for leprosy diagnosis in a hyperendemic area in China
|
Transmissible spongiform encephalopathies ( TSEs ) are caused by the prion , which consists essentially of PrPSc , an aggregated , conformationally modified form of the cellular prion protein ( PrPC ) . Although TSEs can be experimentally transmitted by intracerebral inoculation , most instances of infection in the field occur through extracerebral routes . The epidemics of kuru and variant Creutzfeldt-Jakob disease were caused by dietary exposure to prions , and parenteral administration of prion-contaminated hormones has caused hundreds of iatrogenic TSEs . In all these instances , the development of postexposure prophylaxis relies on understanding of how prions propagate from the site of entry to the brain . While much evidence points to lymphoreticular invasion followed by retrograde transfer through peripheral nerves , prions are present in the blood and may conceivably cross the blood-brain barrier directly . Here we have addressed the role of the blood-brain barrier ( BBB ) in prion disease propagation using Pdgfbret/ret mice which possess a highly permeable BBB . We found that Pdgfbret/ret mice have a similar prion disease incubation time as their littermate controls regardless of the route of prion transmission . These surprising results indicate that BBB permeability is irrelevant to the initiation of prion disease , even when prions are administered parenterally .
Transmissible spongiform encephalopathies ( TSEs ) are progressive , invariably lethal neurodegenerative diseases which include Creutzfeldt–Jakob disease , kuru , fatal familial insomnia and Gerstmann–Sträussler–Scheinker syndrome in humans , scrapie in sheep , and bovine spongiform encephalopathy ( BSE ) in cattle [1] . The infectious agent , termed prion , consists primarily of PrPSc , a conformationally modified form of PrPC , a protein encoded by the gene PRNP [2] . Conversion of PrPC into PrPSc leads to accumulation of insoluble , partially protease-resistant prion protein deposits in the brain parenchyma around neurons and neuronal loss which is accompanied by gliosis and spongiform changes . Deletion of PrPC renders mice resistant to prion infections , indicating that its conversion into PrPC is necessary for the development of disease [1] . Although there have been instances of patients intracerebrally infected by prion-contaminated medical equipment or by dura mater grafts of cadaveric origin [3] , transmission of prion infections occurs more frequently through peripheral routes . The oral route of transmission has caused epidemics of kuru and variant CJD in humans , as well as BSE in cows [3] . Likewise , the parenteral route of prion transmission is highly effective in laboratory mice and hamsters . But how do prions reach the central nervous system ( CNS ) upon entering the body from peripheral sites ? After extraneural inoculation , prions accumulate and replicate in lymphoid tissues [4] . Follicular dendritic cells ( FDC ) and their precursors may constitute the first site of prion amplification [1] . Several studies indicate that prions travel to the CNS along peripheral sympathetic nerves , and the distance between FDC and sympathetic nerve endings specifies the speed of neuroinvasion [5 , 6] . However , none of these findings exclude the possibility that prions , in addition to following the lymph invasive route , may directly colonize the CNS through hematogenic spread followed by direct crossing of the brain vasculature . Prions are present in the blood of hamsters , mice and humans , and it was recently shown that both , PrPc and PrPSc , can cross the blood-brain barrier ( BBB ) [7–9] . Whether this contributes to the initial spread of the disease into the CNS is largely unknown . Here we have addressed the role of the BBB in prion pathogenesis using a genetically modified mouse strain ( Pdgfbret/ret ) which possesses a highly-permeable BBB as the result of the expression of a platelet-derived growth factor B ( PDGF-B ) lacking the PDGF-B retention motif [10 , 11] . We show that Pdgfbret/ret mice succumb to prion disease similarly to their littermate controls regardless of the route of prion transmission . In addition , there are no differences in histopathological characteristics of the disease nor in the resistance to the protease K of PrPSc in the brains of terminally sick Pdgfbret/ret mice compared to the controls . Our study indicates that although PrPSc can cross the BBB [7 , 8] , this route of entry into the CNS is negligible as regards the initiation of the disease when prions are administered intravenously , and highlights the importance of peripheral replication in prion disease pathogenesis in the case of blood-borne transmission .
B6 . 129-Pdgfb<tm3Cbet> [12] heterozygous mice ( Pdgfbwt/ret ) in the C57BL/6J genetic background were crossed to obtain Pdgfbwt/wt , Pdgfbwt/ret and Pdgfbret/ret littermates that were used for prion infection studies . Pdgfbret/ret animals possess an open BBB [10] . The BBB defect occurs at the level of endothelial transcytosis and tracers with a wide range in the molecular weight ( 1 kDa– 200 kDa ) or different chemical composition enter the brain parenchyma in Pdgfbret/ret animals . Animal care and experimental protocols were in accordance with the “Swiss Ethical Principles and Guidelines for Experiments on Animals” , and approved by the Veterinary office of the Canton of Zurich ( permits ZH130/2008 , ZH14/2012 , ZH90/2013 and ZH196/2014 ) . Pdgfbwt/ret and Pdgfbret/ret mice received 2 . 5mg/20g 70 kDa-dextran conjugated to Texas Red ( Invitrogen , Cat # D1864 ) via the tail vein . The tracer was allowed to circulate for 5 hours . Mouse brain tissue was prepared for whole-brain clearing according to published protocols [13 , 14] . Mice were deeply anaesthetized and transcardially perfused with ice cold PBS followed by a fixative mixture of 4% acrylamide , 1% paraformaldehyde , 0 . 05% Bis , 0 . 25% VA-044 in PBS . Mouse brains were removed and post-fixed in the same fixative for 24 hours at 4 °C . The brains were de-gassed , exposed to gaseous nitrogen , and polymerized for 2 . 5 hours at 37 °C . Brains were extracted from the hydrogel and placed in 8% sodium dodecyl sulfate ( SDS ) , 200 mM boric acid , pH 8 . 5 ( clearing solution ) . Brains underwent clearing by electrophoresis ( 4–8 hours ) . Cleared brains were then washed in PBS . The refractive index was equilibrated with refractive-index matching solution prepared according to published protocols [14] . Brains were imaged using a custom mesoscale selective plane illumination microscope ( mesospim . org ) that will be described in detail elsewhere . Images were processed using Image J and Imaris ( Bitplane ) software . Mice were infected with the Rocky Mountain Laboratory ( RML ) scrapie strain ( passage 6 , RML6 ) . Three different inoculation routes were used: intracerebral , intravenous and intraperitoneal . For inoculations , we used 30 μl of RML6 brain homogenate prepared in a solution of 0 . 32 M sucrose containing 5% BSA . Control groups of mice received intracerebrally 30 μl of non-infectious brain homogenate ( NBH , 10% w/v ) prepared from healthy CD-1 mice . Clinical assessment and scoring of mice based on the presence of neurological signs ( including ataxia , kyphosis , priapism , leg paresis , lack of grooming ) was performed as previously described [15] . Mice were euthanized on the day of onset of clinical signs of scrapie according to the approved protocols . One group of mice was inoculated intracerebrally with 30 μl of RML6 brain homogenate containing 1 . 5 log LD50 of infectious agent . Two groups of mice received RML6 intravenously 100 μl of RML6 brain homogenate containing 6 log LD50 and 100 μl of RML6 brain homogenate containing 3 log LD50 of infectious agent . One group of mice received 30 μl of RML6 brain homogenate containing 4 . 5 log LD50 intraperitoneally . Prism software ( www . graphpad . com ) was used to perform statistical analysis . The log-rank test was used to compare the survival curves between Pdgfbwt/wt , Pdgfbret/wt and Pdgfbret/ret littermates . Brains were homogenized in 0 . 32 M sucrose in PBS . Total protein concentration was determined using the bicinchoninic acid assay ( Pierce ) according to manufacturer’s instructions . Samples were adjusted to 1 μg/μl and digested with proteinase K ( PK ) ( 20 μg/μl ) in PBS , 0 . 5% SDS and 0 . 5% NP-40 for 30 minutes at 37 °C . Proteinase K reaction was stopped by adding loading buffer ( Invitrogen ) followed by boiling samples for 5 minutes at 95 °C . PK-treated and untreated samples were separated on a 12% Bis-Tris polyacrylamide gel ( NuPAGE , Invitrogen ) and blotted onto a nitrocellulose membrane . Anti-PrP antibody ( POM1 , 200 ng ml−1 ) [16] was used as a primary antibody which was detected using rabbit anti-mouse IgG1 conjugated to horseradish peroxidase ( HRP ) . Western blots were developed using Luminata Crescendo Western HRP substrate ( Millipore ) and visualized using the FUJI-FILM LAS-3000 system . The glycoform profiles of PrPSc after PK treatment were quantified using the Quantify One software ( BioRad ) . The relative intensity of each PrPSc glycoform ( i . e . di- , mono- , ungylcosylated ) was measured which was expressed then as a percentage of the total signal . Statistical analysis ( two-way ANOVA ) was performed using the Prism software ( www . Graphpad . com ) . Formalin-fixed tissues were treated with concentrated formic acid for 60 minutes at room temperature to inactivate prion infectivity . Tissue was embedded in paraffin and cut into 2 μm sections . After deparaffinization through graded alcohols sections were stained with hematoxylin/eosin . Antibody SAF-84 ( A03208 , 1:200 , SPI-Bio , Waterloo , Australia ) was used to detect partially protease-resistant prion protein deposition on a NEXES immunohistochemistry robot ( Ventana Instruments , Basel Switzerland ) using an IVIEW DAB Detection Kit ( Ventana ) , after incubation with protease 1 ( Ventana ) . Microglia was detected using anti-Iba 1 antibody ( WAKO ) . Sections were deparaffinized through graded alcohols and heat-induced antigen retrieval was performed in citrate buffer ( 0 . 01 M; pH 6 ) . Sections were incubated with anti-Iba1 Ab ( 1∶2500 ) . Stainings were visualized using DAB ( Sigma-Aldrich ) and H2O2 ( Sigma-Aldrich ) , after incubation with a biotinylated secondary antibody ( Vector Laboratories ) followed by the ABC complex solution ( Vector laboratories ) . Sections were counterstained with Hematoxylin . Images of HE and DAB stained sections were acquired using a NanoZoomer scanner ( Hamamatsu Photonics ) and NDPview digital pathology software ( Hamamatsu Photonics ) . The quantification of Iba1 positive cells was performed after DAB immunohistochemistry ( n = 4 fields/mouse , n = 3–6 mice/group ) using Qupath software ( manual quantification function ) [17] . Data were presented as number of Iba1 positive cells/mm2 . Quantification of vacuoles was performed on HE images ( ( n = 8 fields/mouse , n = 3–5 mice/group ) . The algorithm to count vacuoles was developed in MATLAB R2016B using the Image Processing toolbox . Image segmentation was performed using Otsu’s thresholding method ( T = 0 . 7 ) . Only round vacuoles with an area in the range of [200 pixels , 2000 pixels] and with a shape values > 0 . 9 ( where shape was calculated as shape = ( 4*π*Area ) / ( Perimeter^2 ) ) were quantified . The code is available at: https://github . com/AndraCh/Vacuoles_segmentation . Statistical analysis ( one-way ANOVA ) was performed using the Prism software ( www . Graphpad . com ) . PK-resistant PrPSc was quantified in brain tissue homogenate by FRET using monoclonal antibodies POM1 and POM19 [16] . Protein concentration of samples were determined with a bicinchoninic acid assay performed according to the manufacturer’s instructions ( Thermo Fisher Scientific ) . A total amount of 10 μg protein per well was diluted to the correct volume in PBS . To determine PK-resistant PrPSc levels , samples were PK digested using 50 μg/ml PK ( Roche ) at 37°C for 30 min under constant agitation . Digestion was stopped by adding PMSF to a final concentration of 2 . 24 mM and samples were incubated for 10 min at room temperature ( RT ) . Denaturation of remaining PrPSc in samples was achieved by the addition of NaOH to a final concentration of 56 mM and samples were incubated for 10 min at RT under constant agitation . To neutralize the samples , NaH2PO4 was added to a final concentration of 66 mM and incubated for 10 min at RT . Samples were pipetted in triplicates to a 384-well OptiPlate ( Perkin Elmer ) . For FRET assay we used two in-house produced monoclonal antibodies recognizing different epitopes of PrP ( POM19 and POM1 ) [16] . POM19 was coupled to Europium ( EU , FRET donor ) and POM1 was coupled to allophycocyanin ( APC , FRET acceptor ) . The antibody pair was diluted in 1X Lance buffer ( Perkin Elmer ) to a final concentration of EU-POM19 of 2 . 5 nM and APC-POM1 of 5 nM . After adding the antibody pair to samples , the plate was centrifuged at 2000 g for 1 minute and incubated overnight at 4°C . The following day , FRET was measured using a multilabel plate reader ( EnVision , Perkin Elmer ) . The excitation wavelength was 337 nm and emission wavelength for EU was 615 nm and for APC 665 nm . Following measurement , the net-FRET signal was used in accordance with formula published in [18] . For subsequent analysis , the triplicates were averaged and the signal for non-infectious brain homogenate was subtracted to remove the background . PrPSc levels are presented as a ratio of PrPSc in each individual animal to the average of wild-type littermates . Statistical analysis ( one-way ANOVA ) was performed using the Prism software ( www . Graphpad . com ) .
In order to detect any possible role of the BBB in prion pathogenesis we used Pdgfbret/ret mice which express a hypomorphic variant of PDGFB [10] . The cerebrovascular tree in these animals is highly permeable to blood-borne macromolecules , as has been demonstrated in several studies using many orthogonal experimental approaches including immunohistochemistry , spectrophotometry , MRI [10 , 11 , 19] . Surprisingly , these mice are viable and enjoy an almost-normal life span despite their deeply dysfunctional BBB . These findings made it possible to perform prion infection experiments and study the development of disease over many months . Certain areas of the brain are differentially susceptible to neurodegeneration induced by different prion strains [20] . It is conceivable that the BBB defect of Pdgfbret/ret mice may not coincide with the areas of selective vulnerability to prions . Therefore , we investigated the regional characteristics of BBB permeability in Pdgfbret/ret mice using electrophoretic clearing based on the CLARITY method [13] in custom designed clearing chambers . After receiving an intravenous injection of 70 kDa dextran conjugated to TexasRed , mice were perfused , brains were cleared and then imaged using a mesoscale selective-plane illumination microscope . The entire brain of Pdgfbret/ret animals showed passage of 70 kDa-dextran Texas Red into the brain parenchyma ( S1B Fig and S1 Movie ) . The strongest leakage of dextran was detected in the cortex compared to other brain regions , which is in agreement with previously published data [19] . In control animals ( Pdgfbwt/ret ) , no fluorescent signal can be detected in the brain parenchyma ( S1A Fig ) . Thus , these data further demonstrate an enhanced permeability of the BBB in the entire brain of Pdgfbret/ret mice . To test whether Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) are competent for prion replication and succumb into prion disease similarly , mice were inoculated intracerebrally with RML6 . There was no difference in disease incubation between Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) when infected intracerebrally ( Fig 1A ) . Inoculation of brain homogenate can induce autoimmune encephalitis , and it is not known whether such pathology may be exacerbated by a leaky BBB . Therefore , Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) were inoculated with non-infectious brain homogenate prepared from CD-1 strain as a control . None of these mice developed clinical signs of disease . They were sacrificed 350 days post-inoculation and showed no signs of encephalitis or prion pathology ( Figs 2 and 3 and S2 and S5 Figs ) . When prions are administered peripherally , neuroinvasion is dependent on peripheral replication in lymphoid tissues [4] . Therefore we assessed the course of prion disease in Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) after intraperitoneal administration of prions . Similarly to intracerebral inoculation , Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) showed no differences in prion disease pathogenesis after intraperitoneal inoculation ( Fig 1B ) . Thus , we conclude that the altered BBB does not affect prion neuroinvasion when prions are administered peripherally nor lead to accelerated prion pathogenesis when administered intracerebrally . It has been previously claimed that intravenously injected prions reach the CNS within minutes [7] , and the authors of that study concluded that the quantity of PrPSc that reaches the brain via the BBB is sufficient to induce the disease in mice possessing a normal BBB . However , another study with Syrian hamsters found that although prions were found in the CNS few days after peripheral administration , the levels of prions were sub-infectious [21] . The mode of PrPSc transport into the CNS via brain endothelium is not known . The brain vasculature of Pdgfbret/ret mice is permeable to plasma proteins such as albumin , IgG , and the BBB permeability occurs at the level of endothelial transcytosis [10] . Normal brain endothelial cells show a paucity of transcytotic vesicles , however , the increased transcytosis is seen as a first sign of BBB defect in several brain insults ( e . g . ischemic stroke ) [22] . To directly assess the role of the open BBB in prion pathogenesis , Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) received RML6 intravenously , either a high dose ( 6 log LD50 ) or a low dose ( 3 log LD50 ) . We reasoned that Pdgfbret/ret animals should show an earlier onset of prion disease compared to the control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) since their BBB is permeable to plasma proteins . However , this was not the case , Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) animals did not show any statistically significant differences in the speed of prion pathogenesis and in their attack rate , either at high dose ( 6 log LD50 , p = 0 . 37 ) or at low dose ( 3 log LD50 , p = 0 . 86 ) of intravenously administered prions ( Fig 1C and 1D ) . The minimal trend towards earlier lethality of Pdgfbret/ret mice was well within the biological variability expected in this kind of experiments . Our results using a mouse model with a compromised BBB due to increased transcytosis show that the permeability of CNS vasculature has a negligible effect on prion disease transmission into the CNS . This observation is in agreement with two decades of studies on the role of lymphoid organs in prion spread . If prions could enter the brain directly after peripheral inoculation , it would be difficult to understand why mice lacking B-cells [23] or complement components [24] experienced delayed neuroinvasion , and why the distance between follicular dendritic cells and peripheral nerve endings controls the speed of neuroinvasion [6] . Instead , the current results validate a model by which peripherally administered prions first colonize the lymphoid organs , then undergo a phase of clinically silent peripheral replication , and finally achieve neuroinvasion by exploiting peripheral nerves belonging to the sympathetic nervous system [5] . Therefore , prions resemble neurotropic viruses such as rhabdoviruses and herpesviruses that utilize retrograde axonal transport to gain access to the central nervous system , thereby bypassing the need for breaching the BBB . However , the limitation of our study is that the amount of PrPSc in the brain parenchyma shorty after intravenous injection with RML6 homogenate has not been quantified since currently available methods for detecting infectious prions in the brain parenchyma shortly after inoculations are not sufficient for this task . Although Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) did not show differences in the prion disease incubation , they may conceivably differ with respect to the histological and biochemical hallmarks of prion infection . We therefore assessed the extent of spongiosis , the most characteristic feature of TSEs , on hematoxylin-eosin stained brain sections . All prion-inoculated animals , but none of the mice injected with NBH , showed the presence of spongiosis regardless of their genotype ( Fig 2A ) . Quantification of vacuolation showed no difference in number of vacuoles and area covered by vacuoles between Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) ( Fig 2B and 2C ) . Microglia , whose activation in prion disease is considered neuroprotective partly by prion clearance [25 , 26] , express Pdgfb [27] . Although it is unlikely that the lack of Pdgfb retention motif will have a cell-autonomous effect on microglia—since microglia do not express the receptor of Pdgfb—Pdgfrb , we nevertheless assessed microgliosis in terminal stage of the disease . No differences in number of microglia in between Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) was detected ( S3 Fig ) . Finally , we investigated the levels of PrPSc accumulation using immunohistochemistry on formalin-fixed paraffin-embedded tissue sections , Western blotting and fluorescence-resonance energy transfer ( FRET ) . Immunohistochemical analysis showed the presence of partially protease-resistant PrP ( PrPSc ) in the brains of all prion-inoculated mice , regardless of the genotype and of the inoculation route , whereas NBH-injected mice never exhibited any PrPSc deposits ( Fig 3 ) . Interestingly , Pdgfbret/ret mice showed more prominent accumulation of PrPSc along blood vessels than wild-type mice ( S4 Fig ) . All prion-inoculated animals showed PrPSc accumulation in spleen ( S5 Fig ) . Western blotting of partially proteinase K-resistant PrPSc showed similar levels of disease-associated PrP among the three genotypes ( Pdgfbwt/wt , Pdgfbwt/ret , Pdgfbret/ret ) for all inoculation routes and prion doses ( Fig 4A–4D ) . Quantification of PrPSc levels using a FRET assay after PK digestion [18] did not show a difference in PrPSc levels between Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) ( Fig 4E ) . It is conceivable that BBB opening would result in unorthodox prion replication and possibly in a strain shift [28] . Thus , we investigated whether differences in the BBB permeability in Pdgfbret/ret mice lead to alteration in prion strain . As a proxy for strain identification we quantified the ratio of the mono- , di- , and unglycosylated forms of PrPSc . No significant difference between glycoform ratios were found among genotypes ( Fig 4F and 4G ) , suggesting that no shift in prion strains had taken place in Pdgfbret/ret animals . The altered BBB in Pdgfbret/ret mice in chronic inflammatory condition could lead to infiltration of peripheral leukocytes which could modify disease cause . However , other studies have shown that during prion disease there is a minimal recruitment of inflammatory monocytes and mice showing various T cell deficiencies develop clinical signs of prion disease with a comparable temporal dynamic/pattern to that seen in wild-type mice [23 , 29] . The development of prion disease does not differ between Pdgfbret/ret and control mice ( Pdgfbwt/wt , Pdgfbwt/ret ) despite of the breached BBB of Pdgfbret/ret animals after all tested inoculation routes , including intracerebral inoculation . Thus , if there is a component in pathology caused by peripheral leukocytes then the effect on the disease has no detectable influence . In conclusion , these data indicate that when prions are inoculated directly into the blood , the intactness of the BBB has a negligible effect on the incubation time of the disease . This suggests that extracerebrally administered prions do not need to trespass the BBB in order to enter the CNS . Alternatively , one could construe that prions can trespass the neurovascular barriers and colonize the brain through a mechanism that is fundamentally independent of the BBB . There is no factual evidence supporting the latter scenario . A wealth of data accrued in multiple model systems supports the idea that prions , after entering the body from extraneural sites , undergo an early phase of replication in lymphoid organs , which is then followed by the colonization of peripheral nerve endings . It is then through the latter , according to this hypothesis , that prions eventually gain access to the CNS—akin to neurotropic viruses such as rhabdo- and herpesviruses . If prion spread to the CNS were directly hematogenic , one might expect the first site of CNS replication to be solely determined by the differential prion replication competence of select brain areas , rather than by the site of inoculation . In reality , however , the first site of CNS invasion after intraperitoneal inoculation with prions corresponds to the segmental projections of the peripheral nerves to the spinal cord [5 , 6] . Moreover , chemical or immunological sympathectomy suffices to prevent neuroinvasion after intraperitoneal prion inoculation [5] . In the framework of the studies enumerated above , the data reported here add to the conjecture that prion spread from the periphery to the brain does not occur by direct transition across the BBB . Besides their significance for the basic understanding of prion neuroinvasion , these results may be of relevance to the possibility of developing effective post-exposure prophylaxis of prion diseases , which may prevent neurodegeneration even after extraneural infection has already taken place .
|
Prion diseases or transmissible spongiform encephalopathies ( TSEs ) are incurable brain diseases caused by conformational changes in the endogenous prion protein . Prions can be transmitted through contaminated food , surgical instruments and blood . Transmission of prions has caused the kuru epidemic in humans and bovine spongiform encephalopathy in cattle , which in turn caused variant Creutzfeldt-Jakob disease ( CJD ) in humans . Furthermore , injection of prion-contaminated hormones has caused hundreds of TSE cases . In order to develop drugs to prevent the spread of prions into the brain after exposure via food or medical procedures , it is necessary to gain an understanding of how prions propagate from the site of entry to the brain . Prions were shown to reach the spinal cord by traveling along peripheral nerves . However , prions are also found in blood . Although normal brain vessels act as a barrier between the blood and brain , some studies suggested that prions in blood may enter the brain via blood vessels . Here we have tested the latter hypothesis using mice with increased brain blood vessel permeability . We found that these mice are similar to wild-type mice in their susceptibility to prion disease and incubation times after peripheral inoculation . These results suggest that passage of prions through the blood-brain barrier may not be relevant to the development of disease , and imply that any effective post-exposure treatment should rather aim at other rate-limiting steps of prion propagation .
|
[
"Abstract",
"Introduction",
"Material",
"and",
"methods",
"Results",
"and",
"discussion"
] |
[
"group-specific",
"staining",
"medicine",
"and",
"health",
"sciences",
"animal",
"diseases",
"nervous",
"system",
"hematoxylin",
"staining",
"enzymes",
"blood-brain",
"barrier",
"enzymology",
"fluorophotometry",
"animal",
"prion",
"diseases",
"infectious",
"disease",
"control",
"zoology",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"infectious",
"diseases",
"staining",
"fluorescence",
"resonance",
"energy",
"transfer",
"zoonoses",
"proteins",
"spectrophotometry",
"biochemistry",
"anatomy",
"central",
"nervous",
"system",
"biology",
"and",
"life",
"sciences",
"proteases",
"spectrum",
"analysis",
"techniques",
"prion",
"diseases"
] |
2018
|
Prion pathogenesis is unaltered in a mouse strain with a permeable blood-brain barrier
|
The molecular makeup of the offspring of a dividing cell gradually becomes phenotypically decorrelated from the parent cell by noise and regulatory mechanisms that amplify phenotypic heterogeneity . Such regulatory mechanisms form networks that contain thresholds between phenotypes . Populations of cells can be poised near the threshold so that a subset of the population probabilistically undergoes the phenotypic transition . We sought to characterize the diversity of bacterial populations around a growth-modulating threshold via analysis of the effect of non-genetic inheritance , similar to conditions that create antibiotic-tolerant persister cells and other examples of bet hedging . Using simulations and experimental lineage data in Escherichia coli , we present evidence that regulation of growth amplifies the dependence of growth arrest on cellular lineage , causing clusters of related cells undergo growth arrest in certain conditions . Our simulations predict that lineage correlations and the sensitivity of growth to changes in toxin levels coincide in a critical regime . Below the critical regime , the sizes of related growth arrested clusters are distributed exponentially , while in the critical regime clusters sizes are more likely to become large . Furthermore , phenotypic diversity can be nearly as high as possible near the critical regime , but for most parameter values it falls far below the theoretical limit . We conclude that lineage information is indispensable for understanding regulation of cellular growth .
The process of cellular growth is both the distinguishing feature of living matter and central to the roles of regulatory networks from microbes to metazoa . Growth and division is also a primary source of phenotypic diversification . For instance , when a bacterial cell divides , and its cellular contents become partitioned into two daughter cells , diffusible cytoplasmic components are often randomly distributed into the daughter cells in a binomial distribution . Such phenotypic diversification permits populations to be robust to unpredictably changing environments , a phenomenon known as bet-hedging . A striking example of this effect is the regulation of growth rate by toxins . Most of the molecular content in the bacterial cytoplasm undergoes growth-mediated dilution ( in some cases , such as most proteins , as the primary mechanism of degradation ) . Reduction in cellular growth rate by a cytoplasmic toxin , or other molecule with toxic effect , creates an effective positive feedback loop , trapping some cells in a growth arrested state until they can escape in changed conditions [1–3] . This mechanism is associated with antibiotic-tolerant persister cells arising in the population , which cause difficulty in antibiotic treatment [4] . Various feedback mechanisms are associated with growth bistability [5] . Thus , understanding the processes that result in growth diversification is an important goal on the path to solving the impending antibiotic resistance crisis . Growth arrested cells typically represent a small subset of a bacterial population [6] . In E . coli , growth arrested persister cells are associated with alterations in metabolic activity via the stringent response [7 , 8] , and with efflux of antibiotics [9] . Depending on the mechanism of induction , persister cell fractions can be spontaneously produced or respond to external stresses [6] . Persistence in E . coli is associated with toxin-antitoxin systems and global metabolic regulation [10] , with a core mechanism of toxins that are neutralized by antitoxins [11] ( Fig 1A and 1B ) . The competing effects of toxin and antitoxin create a threshold in a stoichiometric effect via molecular titration that can cause conditional cooperativity of TA gene regulation [12 , 13] . When accounting for gene expression noise and proteolysis of antitoxins , free toxin levels will gain sufficient concentration to result in a growth feedback mechanism that ultimately induces growth arrest in above-threshold cells . The result is skewed phenotypic distributions , with a core fast-growing group of cells along with rarer , growth arrested cells , as opposed to regression to mean levels observed in networks without the growth arrest threshold ( Fig 1C and 1D ) . Motivated by observations on phenotypic inheritance [14–16] and the effects of lineage correlations on daughter cell phenotypes [17–21] , we asked how much phenotypic diversity could be attained for various levels of endogenous growth regulation , and to what extent lineage determines phenotypic outcomes . Based on our previous study [17] , we hypothesized that a higher chance of growth arrest amplifies the effects of cellular lineage on phenotypic correlations . To explore this hypothesis , we used an established experimental model of threshold-based growth arrest in E . coli to experimentally confirm lineage dependence . We then created a minimal multiscale computational framework that allowed more extensive characterization of the various growth regimes than were possible with time-lapse microscopy . Our computational model represents the processes of cellular growth and division , with binomially distributed inheritance of a simplified toxin-antitoxin-like system subject to stochastic molecular kinetics in individual cells over time . We modeled a functional dependence of growth on toxin concentrations as an exponential function with a key parameter , α , that quantifies how toxic the toxin is . We used various specific realizations of the framework to simulate growth of small bacterial populations from a single common ancestor and growth regulation by the simulated toxin for various toxin:antitoxin production ratios . Our computational results confirm and extend the experimental results , showing that the bet-hedging regime results in complex lineage structures . These results show , for the first time , how important lineage is to growth regulation and bet-hedging phenotypes involving growth . Consideration of lineage is now indispensable for studies on phenotypic heterogeneity , phenotypic memory , and regulation of the growth arrest transition . Finally , our results suggest that lineage space used in evolutionary [22] and multicellular organism development studies [23] is an important concept to apply in studies of bacterial phenotype .
We first sought to establish an empirical basis for growth arrest kinetics and threshold-based amplification of lineage correlations . An established experimental model of threshold-based growth arrest [17] provided a simple way to track growth in a lactose-sensitive strain of E . coli . In this model , lactose stimulates growth at sufficiently low concentrations , but creates toxicity in a subset of cells at high concentration that results growth arrest or death of those cells . Presently , the precise mechanism of toxicity is not known in this model , but the competing effects of lactose import rate and processing rate are the most likely culprit , and the threshold-based mechanism for growth arrest and persistence is established [17] . In the high-lactose condition , bacterial colonies have a slow net growth rate and a high likelihood of any individual cell eventually undergoing growth arrest and/or death . We used time-lapse fluorescence microscopy to track individual microcolonies in a microfluidic device with constant perfusion of fresh minimal medium containing defined concentrations of a single sugar as the sole carbon source . We used two carbon sources: a growth-arrest-prone condition with a high lactose concentration ( 50 g/l ) , and a condition that does not induce a growth arrest threshold , with a moderate glucose concentration ( 2 g/l ) ( Fig 2; S1 and S2 Videos ) . As inferred from extension of cellular major axis length , cells grow exponentially at heterogeneous rates ( Figs 2A and 2B , 2E–2R and S1 ) and are capable of quickly shifting between growth rates , e . g . , from fast to slower or non-growing ( Fig 2B and 2F ) . To identify cases of mid-cell cycle shifts in growth rate , we fit each cell cycle to an exponential growth model , applied Bonferroni correction to the resulting fit significance levels , and selected the non-significant cases ( S3 Fig ) . A constitutive fluorescent reporter provides clear visual evidence of mother-daughter cell correlations only in the growth arrest-prone condition ( Fig 2C and 2G ) . We reconstructed the microcolony lineage in both conditions to quantify the effects of non-genetic inheritance in this experiment ( Fig 2D and 2H ) . The result of the growth arrest threshold is a striking effect on the structure of the lineages . The growth arrest-prone lineage shows distinct clusters of growth arrested or dead cells , and clusters of faster growing cells , resulting in an asymmetric tree ( Fig 2D ) . On the other hand , absent the growth arrest threshold , the tree is nearly symmetric ( Fig 2H ) . In the growth arrest prone condition , we classified cells into being growth arrested or dead ( apparent growth rate = 0 ) or actively growing . Of the 63 total cells in the final lineage , 16 ( 25 . 4% ) were determined to be completely growth arrested or dead at the final time point . We determined the pairwise lineage distance , defined as the time since the most recent common ancestor , for three subsets: all cells , only growing cells , and only growth arrested cells ( S2 Fig ) . The all-growing and all-growth arrested subsets both had significantly closer lineage distances compared to the all cells set ( p < 0 . 05 , Mann-Whitney U ) . From these results , we conclude that lineage has a strong effect on phenotypic heterogeneity during colony development around a growth-modulating threshold . To determine the minimal set of mechanisms necessary to reproduce the interactions between threshold-based molecular regulation of growth rate and population dynamics , we created a computational model containing cell agents growing and dividing at a typical rate for enteric bacteria ( 30 minute doubling time ) , each with a cell volume and division upon doubling of the volume . Each cell agent has embedded stochastic kinetics of a growth-inhibiting molecule ( analogous to a toxin ) and a neutralizing molecule that binds and prevents toxicity ( analogous to an antitoxin ) . As discussed in more detail in Methods , we assume toxin and antitoxin production , growth-mediated dilution , and binding-unbinding kinetics of the molecules . We used a phenomenological exponential function layer that translates between concentrations of toxin and resultant growth rate , with a single parameter , α , that determines the level of toxicity . The key similarity between our experimental and computational approaches is the existence of a threshold in the molecular network that determines the growth rate of the cell . There are many potential mechanisms for such a threshold to arise , as discussed in the Introduction . We do not claim that the mechanism implemented in the computational model is the same as the experimental model . Rather , there is an underlying fundamental interplay between growth regulation and lineage structure that we will show is conserved . To determine the effect of the growth threshold on microcolony dynamics , we scanned the rate of toxin production , keeping antitoxin production constant . ( In most natural toxin-antitoxin systems , the antitoxin is unstable . We simulated this case as well , below ) . The simulations were seeded with a single cell growing with excess antitoxin and permitted to grow for 100 simulation minutes before changing the toxin production rate to a positive value . After several generations of growth , we found three qualitative regimes across different toxin production rates: symmetrical growth with no or little growth arrest ( toxin production rate 0–2 . 5 /min ) , a critical regime with clusters of growing and growth arrested cells ( toxin production rate 3–4 . 5 /min ) , and a regime of nearly instantaneous growth arrest ( toxin production rate >4 . 5 /min ) with the colony trapped in its near-initial state . Fig 3 shows representative cases with growth rate ( Fig 3A ) or toxin concentration ( Fig 3B ) depicted with coloring of each cell . Sub-lineages of fast-growing and slow-growing cells are evident in the critical regime ( with toxin production rate 5–6 /min; Fig 3A ) . Lineage effects are also evident from toxin levels , where there are sublineages escaping from entry into high toxin concentrations ( blue clusters in Fig 3B ) . The precise time of entry into growth arrest can have a large effect on toxin levels , suggesting that growth rate is a more precise phenotype to follow for the study of lineage effects in this system . To quantitatively characterize the properties of growth transitions in our simple computational framework , we considered the fate of simulated microcolonies at 250 minutes of growth , which is shortly before the fastest growing cases begin to become computationally intractable , but after the population size is beyond the minimal requirement to be considered a microcolony . Mean population growth rates and toxin concentrations across multiple ( N = 100 ) replicates reveal a growth-regulatable region flanked by regions of almost full growth and almost complete growth arrest ( Fig 4A ) . In the region where population growth is low but positive , toxin concentrations increase monotonically but non-linearly with increases in toxin production ( Fig 4A ) . To quantify the amount of lineage information shared by pairs of cells in their phenotypes , we calculated mutual information between phenotypic differences between pairs of cells and pairwise lineage distance . From each simulation , we sampled one pair of cells randomly to ensure independent , identically distributed samples and performed a resampling procedure 100 times to increase the confidence in our estimate . This was done for absolute growth rate differences and absolute toxin concentration differences ( Fig 4B ) . Various studies of have found mutual information between different points on a lattice to be indicative of a phase transition [24 , 25] . While our model may not exhibit a true phase transition , our mutual information estimator reveals a similar peak for both growth rate and toxin concentrations in the critical regime , where the population growth rate is most sensitive to changes in toxin production rates . Distributions of growth rates reveal the underlying population structure not evident from mean growth rates shown in Fig 4A . Distributions that emerge from the model include uniformly fast ( Fig 4C , top left in Fig 4D ) or slow growing ( Fig 4C , bottom right in Fig 4D ) , heterogeneous between fast and slow growing ( top right and bottom left in Fig 4D ) , and long-tailed with a peak at slow growth ( bottom left in Fig 4D ) . To examine a further indicator of criticality in this system , we calculated the dynamics of growing cell numbers below ( toxin production rate 0–2 . 5 /min ) , near ( toxin production rate 3–4 . 5 /min ) , and above the regulatable region ( toxin production rate >4 . 5 /min ) of growth rate . With toxin production well below the regulatable region , the predicted cell growth becomes equivalent to an uncoupled case where toxin has no effect on growth . Growing cell numbers show variability between simulation replicates near the critical region ( Fig 5A ) . Over time , the dynamics of the mean number of growing cells approaches exponential growth at low toxin production rates , critical growth at intermediate toxin production rates ( as shown in Fig 5A ) , and extinction ( elimination of all growth ) at high toxin production rates . Mean cell numbers in critical growth show persistent oscillations that dampen as the simulated growth rates become decorrelated by noise ( Fig 5A ) . As toxin production approaches the critical regime , some cells accumulate high toxin and , depending on individual cellular toxin accumulation , subsets of the population will enter the exponential or extinction phase . Thus , the time required to conform to the exponential or extinction regimes is high in the critical regime , reminiscent longer relaxation times observed near critical points in other models [e . g . 26] . Autocorrelations of growing cell numbers at lag times after the onset of toxin production reveal this effect . For example , high autocorrelation around lag time 100 min in critical regime ( vertical dotted line ) signifies growth remaining correlated for a longer time compared to the autocorrelation at toxin production rate 3 . 0 /min . The presence of more than two zeroes in the absolute autocorrelations indicates the oscillatory regime ( Fig 5B ) . If lineage is capable of constraining the attainable phenotypes of offspring cells , it stands to reason that the amount of phenotypic heterogeneity attainable in a microcolony is lowered by lineage dependence in systems that generate heterogeneity by diversifying growth rates . It is difficult to generalize what constitutes meaningful diversity in growth rates; small changes may or may not be important to fitness in the long run , but the importance of the distinction between growth arrested and fast-growing cells is clear . Therefore , we used two possible definitions of meaningful diversity: in one , arbitrarily small changes in growth rate or toxin concentration are meaningful . In the other extreme , we assumed that only growing versus non-growing cells ( or high versus low toxin ) is a meaningful distinction . We quantified the phenotypic heterogeneity as information entropy ( base 2 ) , binning the simulated cells according to the two definitions of diversity ( Fig 6 ) . We calculated the maximum entropy in the fine-grained binning case by assuming each cell had a unique value . Note that the maximum entropy is extensive , decreasing with lower total cell count ( Fig 6A ) . In the binary case , the maximum entropy is simply 1 bit . Regardless of the definition used , the peak entropy of the population can get surprisingly close to the maximum entropy . Note that peak entropy of growth rate nearly coincides with peak mutual information between growth rate differences and lineage distance ( Fig 6 , vertical line ) . However , entropy away from this peak sharply decreases from the maximum . In the critical regime , population heterogeneity is affected by two key factors: sensitivity of growth rate to toxin and lineage dependence . Given that we observed higher lineage dependence in the critical regime , the key question here is whether this dependence reduces the possible attainable heterogeneity in bet-hedging . The entropy plot ( Fig 6 ) shows that sensitivity of growth rate to toxin dominates and thus phenotypic heterogeneity is maximal at when the lineage is most structured . To explore the generality of our results , we created models with variations on the original , and tested for lineage dependence . The first set of variations test two simplifications in the primary model: stability of the antitoxin , and bursty production of the molecular species . While we regard the model to be a general threshold-based growth control mechanism , it is worthwhile to determine if a toxin-antitoxin module with unstable antitoxin qualitatively reproduces our main results . Varying the stability of the antitoxin , we indeed found the same qualitative results ( S4A Fig ) . Similarly , simulating bursts of gene expression producing toxin and antitoxin produced the same qualitative results ( S4B Fig ) . Our next model variation was to vary the effect of growth regulation , increasing it ( α = 0 . 3 in g ( T , t ) ; see Methods below ) and abolishing it completely ( α = 0 in g ( T , t ) ) . As expected , a larger quantitative effect of toxin preserved the main results , but shifted the toxin concentration necessary to see the lineage dependence ( S4C Fig ) . Abolishing growth regulation eliminated the peak in mutual information , and thus lineage dependence ( S4D Fig ) . Large clusters of growth arrested cells could have effects on the spatial development of bacterial colonies , as daughter cells tend to be correlated in space as well . We therefore asked what growth arrested cluster size distributions arise in the region where there is high mutual information between growth rate and lineage distance . We performed 10 , 000 simulations each and clustered the end-point populations according to lineage neighbors having similar growth rate ( with a cutoff of 0 . 01 /h to be considered growth arrested ) . Resulting clusters were pooled across simulations of the same parameter set . We present distributions of raw absolute cluster size , not normalized . Below the critical regime , the absolute cluster size distribution is nearly exponential ( Fig 7 , red line with exponential fit as gray dashed line ) . As the probability of growth arrest increases ( with high toxin production rate ) , the distributions diverge from exponential to make large clusters of growth arrested cells more likely ( Fig 7 ) . At higher toxin production rates , the distribution is bimodal between large clusters and single growth arrested cells .
Regulation of growth is a central part of phenotypic control . Many factors can control growth rate , including extrinsic conditions such as starvation , and intrinsic regulators of growth that often operate with a threshold-based mechanism . Using an experimental model of threshold-based growth arrest arising from metabolic toxicity , we tracked cell growth in a bacterial microcolony with a high probability of undergoing the growth arrest transition , and a colony grown in a condition that does not display the threshold-based growth arrest . We found several large , discrete shifts in growth rate to occur at a faster timescale than our 5-minute recording intervals ( Fig 2 ) . Quantifying the lineage dependence of cellular growth phenotype , we found that growth arrested or dead cells tend to be clustered in the lineage , as do fast-growing cells . The difference in lineage shapes between the growth arrest-prone and non-growth arrest prone conditions is striking ( Fig 2D and 2H ) . We therefore sought the simplest possible model of microcolony growth dynamics that reproduces the effect . Our basic model captures single-cell biochemical kinetics on one scale ( microscopic ) interfacing population growth dynamics on another scale ( macroscopic ) . We found striking phenotypic lineage dependence to emerge with the following criteria: ( i ) growth rate dependence on a toxin; ( ii ) stochastic dynamics around a cellular threshold embedded within the network; ( iii ) kinetic parameters calibrated so that the population average growth rate is near the regulatable region . As the probability of cellular transition to growth arrest increases , the mutual information between growth rate and lineage distance increases to a peak , then decreases as the simulated microcolony reaches the condition of immediate growth arrest . This transition bears a resemblance to a phase transition , with correlation of microscopic length scales peaking at the critical boundary . Here , the correlation length is in lineage space: we have assumed no traditional spatial information about the cells in the simulation . Lineage space is a binary tree growing with extinction probability based on microscopic dynamics . Distances are modified by dynamical growth rates , which explains why a higher probability of heterogeneous growth results in structured trees . Thus , relating persister and other threshold-based growth arrest mechanisms to the established mathematics of branching processes [27 , 28] is an important direction for microbial physiology . After 100 simulated minutes we imposed a continuous rate of increased toxin production ( or antitoxin degradation , in one derived model ) on the developing microcolony . The constant input of more toxin created an irreversible threshold . Once a cell crosses the growth arrest threshold , there is an irreversible stoppage of growth that arises from toxin growth feedback . The growth arrest condition can then be considered an absorbing state . Continuous transitions from active to absorbing states are generically characterized by the scaling properties of critical directed percolation [29–31] . Our model qualitatively reproduces characteristics of directed percolation , including longer relaxation times near the critical region ( Fig 5 ) and different regimes of growth arrested cluster size distribution ( Fig 7 ) . However , the dimensionality of the space is unclear , and may be shaped by the probability of growth arrest . Thus , we are doubtful that bet hedging quantitatively conforms to the classic criteria for directed percolation . If lineages impart spatial structure onto growth phenotypes , then do they impose an upper limit to the level of phenotypic heterogeneity that can be attained by a microcolony ? The population is most sensitive to fluctuations directly in the region with the highest lineage dependence , the latter of which appears to imply a dampening of phenotypic heterogeneity . However , multiple methods of measuring total population entropy suggest that the population can still approach the maximum total entropy in cases where growth rates are both finely-binned and binned into only two phenotypes–growing and growth arrested ( Fig 6 ) . Heterogeneity is reduced as the population reaches either extreme of high or low average toxin level . Thus , counterintuitively , a more highly structured lineage yields a higher level of heterogeneity . Lineage plays an interesting role in determining the phenotypes of extant growing cells , but it does not appear to restrict what phenotypes can be attained . The purely intracellular phenomena considered here allow lineage to be the only type of space considered . However , closely related cells in many conditions , such as surface-attached conditions or channels , will be physically closer together as well . In many bacterial colonies with a substantial chance of endogenous and exogenous conditions interacting to determine the growth arrest transition ( such as quorum sensing ) , an information metric that includes components of both real space and lineage space will need to be considered .
E . coli B REL606 lacI−PlacO1-GFP was grown from -80° C cryogenic culture for 18 h in LB medium in a shaking incubator ( 37° C ) , acclimatized by incubating in Davis minimal medium containing either 50 mg/ml lactose ( DMlac50 ) or 2 mg/ml glucose ( DMglc2 ) for 24 h , and resuspended either in fresh DMlac50 or DMglc2 culture , respectively , for 3 hours before beginning time-lapse microscopy . We used an Olympus IX81 inverted fluorescence microscope with an incubated imaging chamber ( Olympus , Tokyo , Japan ) . The chamber with objective was pre-heated , bacterial cultures were added to a pre-heated CellAsic ONIX microfluidic plate ( Millipore , Billerica , Massachusetts ) at an approximate OD450 of 0 . 005 , and a continuous media flow of 1 psi DMlac50 or DMglc2 was maintained for the duration of the experiment . Images in brightfield and green fluorescence ( 488 nm stimulation / 509 nm emission ) channels were captured every 5 minutes with a 4k CMOS camera , followed by ZDC autofocus . For the DMlac50 experiment , we used a 100x oil immersion objective . Due to technical issues with the objective , we used a 60x air objective for the DMglc2 experiment . Thus , the pixel lengths of the cells between the two experiments should not be directly compared . Images were cropped after identifying a stable microcolony originating from a single cell . We developed a semi-supervised cell tracking algorithm in Mathematica ( Wolfram Research , Champaign , Illinois ) with manually input cell division times and cell lengths . From this information , we reconstructed the lineage and approximated growth rates with exponential growth models . When mapping the growth rates to the lineages in Fig 2 , we approximated growth rates of cells with non-significant exponential fits using piecewise linear regression as reviewed in [32] . To capture the minimal mechanisms necessary that recapitulate non-genetic inheritance and effects of cellular lineage , we created a multiscale growth simulation framework with individual cell agents , each containing a molecular network of interacting proteins , referred to as toxin and antitoxin , with toxin affecting cellular growth rate . We track the simulated number of toxin and antitoxin molecules as well as cell volumes for each cell agent across time . In the next time step , t+δt , the number of toxin and antitoxin molecules are determined by stochastic simulation ( below ) and are updated for that cell . Cellular growth rates are set by a deterministic function of the toxin concentration ( #/vol ) . The change in the volume ( δv ) in δt is determined by the amount of toxin present at that time . When cell volume doubles , the number of each molecule is distributed binomially into the two daughter cells . From that time on , the two daughter cells are labeled as different cells and are iterated in the same way . We initiate each simulation as a single cell with no toxin and allow growth for a few generations ( 100 minutes ) before applying toxin production rate ( or antitoxin degradation rate ) of a given quantity . The primary purpose of this model is to capture the qualitative effect of the growth arrest threshold , so several important details about the biophysics of kinetics in growing cells were omitted , such as the effects of chromosome replication and the volume dependence of bimolecular stochastic reaction propensities . We sought to develop a sampling methodology to ensure independent , identically distributed samples from lineage simulations to estimate the mutual information between lineage distance d and phenotypic differences between pairs of cells φ . Phenotypic differences ( φ ) could be growth rate or intracellular toxin concentration . To do so , we performed 100 independent simulations in each condition , and randomly drew a single pair of cells from each lineage . Our estimate of mutual information was calculated from the resulting distribution of i . i . d . samples: I ( D , Φ ) =∑φ∈Φ∑d∈Dp ( d , φ ) log2 ( p ( d , φ ) p ( d ) p ( φ ) ) . A more accurate estimate of absolute mutual information may extrapolate to an infinite sample size . In our case , the relative mutual information between different locations in parameter space suffices to demonstrate the existence of a strong lineage dependence for certain parameter ranges . To estimate the uncertainty of our relative mutual information estimate , we resampled 100 cell pairs with replacement and present the resulting mean ± standard deviation . Entropy was calculated by H ( X ) =−∑i=1np ( xi ) log2p ( xi ) , where p ( xi ) represents the probability mass function of a discrete variable X . X could be growth rate or toxin concentration . We considered a simple network consisting of three variables: toxin , antitoxin and toxin-antitoxin bound complex . Possible reaction events are synthesis of toxin and antitoxin , and binding and unbinding between toxin and antitoxin molecules . The reaction scheme for the basic model is: →ktT→g ( T , t ) →kaA→g ( T , t ) T+Akb⇌kuTA→g ( T , t ) The parameter kt is the toxin production rate varied in the simulations . Antitoxin production parameter , ka , is kept constant ( ka = 4 . 2 /min ) to allow the production ratio of toxin and antitoxin to be changed . Growth-mediated loss is implemented through g ( T , t ) which is a function of the cell volume in the algorithm ( below ) . Parameters kb and ku are binding and unbinding rates; kb = 0 . 1 and ku = 0 . 1 throughout . In the most basic model , each species is considered long-lived on the timescales of the simulation , so we do not consider any additional degradation processes . Variations on this model are discussed in Results . The relationship between toxin concentration and cellular growth rate , the most phenomenological part of the framework , captures the interface between molecular and population dynamic scales . We reasoned that , while some random factors may reduce or increase the effect of toxin , the generality with which toxin affects global protein synthesis rates [11 , 33–37] means that many stochastic effects will cancel , resulting in a nearly deterministic relationship . Because toxin levels generally halt ongoing processes without significant delay [38–41] , we approximated the effect of a given toxin level to be instantaneous . This assumption is supported by our experimental results , which show shifts in growth rate faster than the 5 minute intervals measured ( Fig 2 ) . We thus constructed a deterministic function to reflect the functional dependence of growth on toxin concentrations: g ( T , t ) = λe-αT ( t ) /Ω ( t ) , where α is a parameter that represents the toxicity of the toxin , T . We used α = 0 , 0 . 1 and 0 . 3 to represent cases with no toxicity , moderate toxicity , and high toxicity , respectively . Python scripts are given in S1–S3 Models . To illustrate the effects of growth arrest on distributions of growth-modulating cytoplasmic contents ( Fig 1 ) , we created a simplified computational model with constant production , constant sub-threshold generation times , and binomially distributed molecular contents between two daughter cells . One simulation for each initial condition was run for 12 generations , with 10 molecules produced per generation , and a growth arrest threshold of 20 molecules . Initial conditions were 0 , 10 , 20 , or 30 molecules . A second case with no threshold was simulated with the same parameters and initial conditions . The Mathematica code is given in S4 Model . The exact functional dependency of growth on toxin is unknown . In our stochastic simulation framework , we considered an exponential dependence of growth on toxin . Fig 1B depicts a deterministic model of toxin growth feedback by a free toxin as follows: T˙=kt−γθθ+T , where kt is the toxin production rate , γ is the maximum growth rate , and θ determines the toxicity level of the toxin . We chose the Hill form for the deterministic model because it has a closed-form steady state . The steady state is T^=ktγ−ktθ . When kt/θ > γ , there is no steady state at this scale and the containing cell is expected to enter growth arrest . This simple model demonstrates the basis for growth feedback-induced growth arrest in a single cell . For Fig 1B , parameters are: kt = 4 . 2 /min , γ = 0 . 023 , and θ = 100 molecules . We note that the basic growth arrest threshold effect readily emerges in both Hill and exponential model forms , and likely a variety of other mathematical forms .
|
One of the most important characteristics of a cell is whether it is growing . Actively growing cells can multiply exponentially . In the case of infections and cancer , growth causes problems for the host organism . On the other hand , cells that have stopped growing can allocate cellular resources toward different activities , such as bacteria surviving antibiotics and tissues in multicellular organisms performing their physiological roles . Observing small bacterial colonies in a microscope over time , we have found that cells closely related to each other often have similar growth state . We were curious if lineage dependence was an intrinsic property of growth regulation or if other factors were needed to explain this effect . We therefore built a computational model of a growing and dividing cellular colony with an encoded growth regulation network . We found that regulation of growth is sufficient for lineage dependence to emerge . We next asked if lineage dependence constrains how diverse the cellular population can become . We found that cellular diversity can reach a peak that is nearly as high as possible near the conditions that have the highest lineage dependence , but that most conditions do not permit such high diversity . We conclude that lineage is an important constraint and discuss how the growth arrest transition is in some ways like a phase transition from physics , and in some ways strikingly different , making it a unique phenomenon .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"disaccharides",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"carbohydrates",
"organic",
"compounds",
"toxicology",
"toxic",
"agents",
"simulation",
"and",
"modeling",
"toxicity",
"thermodynamics",
"research",
"and",
"analysis",
"methods",
"ecological",
"metrics",
"antitoxins",
"entropy",
"chemistry",
"species",
"diversity",
"physics",
"cell",
"biology",
"ecology",
"organic",
"chemistry",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"lactose"
] |
2018
|
Lineage space and the propensity of bacterial cells to undergo growth transitions
|
Bacterial pathogens causing systemic disease commonly evolve from organisms associated with localized infections but differ from their close relatives in their ability to overcome mucosal barriers by mechanisms that remain incompletely understood . Here we investigated whether acquisition of a regulatory gene , tviA , contributed to the ability of Salmonella enterica serotype Typhi to disseminate from the intestine to systemic sites of infection during typhoid fever . To study the consequences of acquiring a new regulator by horizontal gene transfer , tviA was introduced into the chromosome of S . enterica serotype Typhimurium , a closely related pathogen causing a localized gastrointestinal infection in immunocompetent individuals . TviA repressed expression of flagellin , a pathogen associated molecular pattern ( PAMP ) , when bacteria were grown at osmotic conditions encountered in tissue , but not at higher osmolarity present in the intestinal lumen . TviA-mediated flagellin repression enabled bacteria to evade sentinel functions of human model epithelia and resulted in increased bacterial dissemination to the spleen in a chicken model . Collectively , our data point to PAMP repression as a novel pathogenic mechanism to overcome the mucosal barrier through innate immune evasion .
Epithelial barriers form a first line of defense against microbial invasion . However , the ability to cross this physical barrier does not automatically result in systemic dissemination of the invading microbe . For example , non-typhoidal Salmonella serotypes , such as Salmonella enterica serotype Typhimurium ( S . Typhimurium ) , invade the intestinal epithelium using the invasion associated type III secretion system ( T3SS-1 ) [1] and employ a second type III secretion system ( T3SS-2 ) to survive within tissue macrophages [2] . Despite the ability of non-typhoidal Salmonella serotypes to penetrate the epithelium and survive in macrophages , the infection remains localized to the terminal ileum , colon and mesenteric lymph node in immunocompetent individuals [3] . S . enterica serotype Typhi ( S . Typhi ) differs from non-typhoidal serotypes by its ability to cause a severe systemic infection in immunocompetent individuals termed typhoid fever [4] . However , little is known about the virulence mechanisms that enabled S . Typhi to overcome mucosal barrier functions and spread systemically , which is at least in part due to the lack of animal models for this strictly human adapted pathogen . The chromosomes of Salmonella serotypes exhibit a high degree of synteny , which is interrupted by small insertions or deletions . One such insertion in S . Typhi is a 134 kb DNA region , termed Salmonella pathogenicity island ( SPI ) 7 , which is absent from the S . Typhimurium genome and likely originates from a horizontal gene transfer event , as indicated by the presence of flanking tRNA genes [5] . Within SPI 7 lies a 14 kb DNA region , termed the viaB locus [6] , which contains genes required for the regulation ( tviA ) , the biosynthesis ( tviBCDE ) , and the export ( vexABCDE ) of the Vi capsular antigen [7] . In addition to activating expression of the S . Typhi-specific Vi capsular antigen , the TviA protein represses important virulence factors that are highly conserved within the genus Salmonella . These include genes encoding flagella and T3SS-1 , whose expression in S . Typhi is reduced by a TviA-mediated repression of the master regulator FlhDC [8] . However , the consequences of these changes in gene regulation for host pathogen interaction remain unclear . Here we addressed the biological significance of TviA-mediated gene regulation . To explore how acquisition of a new regulatory protein impacted host microbe interaction , we determined whether introduction of the tviA gene into S . Typhimurium resulted in similar changes in gene expression as observed in S . Typhi . We then investigated how these TviA-mediated changes in gene expression affected the outcome of host interaction in an animal model , the chicken , in which S . Typhimurium causes a localized enteric infection .
In S . Typhi , TviA-regulated genes have been identified and encompass the flagella regulon and genes encoding T3SS-1 [8] . To determine how TviA affects gene expression in a non-typhoidal serotype , the tviA gene was introduced into the S . Typhimurium chromosome and the gene expression profile compared to a published gene expression profile of TviA-regulated genes in S . Typhi [8] . Cluster analysis of gene expression profiles revealed that TviA influenced the transcription of similar regulatory circuits in S . Typhimurium and S . Typhi ( Figure S1 ) , including genes encoding regulatory , structural and effector components of the T3SS-1 , and genes involved in chemotaxis , flagellar regulation and flagellar biosynthesis . To validate results obtained from gene expression profiling , relative transcription levels of genes encoding the flagellar regulator FlhD , the flagellar basal body protein FlgB , the flagellin FliC , and the T3SS-1 regulator HilA were determined in both serotypes by real-time qRT-PCR ( Figure 1 ) . Strains lacking the tviA gene ( i . e . the S . Typhimurium wild-type strain , the S . Typhimurium ΔphoN mutant and the S . Typhi ΔviaB mutant ) contained significantly higher mRNA levels of hilA , flhD , flgB , and fliC than observed in strains carrying the tviA gene ( i . e . the S . Typhi wild-type strain , the S . Typhi ΔtviB-vexE mutant and the S . Typhimurium ΔphoN::tviA mutant , a strain in which the phoN gene had been replaced by the tviA gene ) . Expression of the flagellum is controlled by the master regulator FlhDC ( reviewed in [9] ) and is reduced under low osmolarity in S . Typhi compared to S . Typhimurium [10] . Osmoregulation is mediated through the EnvZ/OmpR system in S . Typhi , which controls the availability of TviA . Under conditions of low osmolarity , TviA is expressed and represses flhDC transcription , thereby negatively regulating flagella biosynthesis [8] , [11] . To understand the consequences of acquiring tviA by horizontal gene transfer , we determined whether differences in flhDC transcription between S . Typhi and S . Typhimurium could be fully accounted for by TviA-mediated gene regulation . Therefore , expression of flhC in S . Typhi and S . Typhimurium was monitored using transcriptional fusions to the Escherichia coli lacZYA reporter genes ( Figure 2 ) . In the S . Typhi wild-type strain , flhC expression increased with increasing salt concentrations present in the culture medium ( Figure 2A , dark gray bars ) . The S . Typhimurium wild-type strain exhibited a strikingly different flhC gene expression pattern , which peaked at medium salt concentrations ( between 0 . 1 and 0 . 2 M NaCl ) ( Figure 2A , light gray bars ) . Removal of the tviA gene in the S . Typhi ΔviaB mutant resulted in an flhC gene expression pattern ( Figure 2A , open bars ) that was similar to that of the S . Typhimurium wild-type strain . Similarly , introduction of tviA into S . Typhimurium resulted in a flhC gene expression pattern ( Figure 2A , closed bars ) resembling that of the S . Typhi wild-type strain . TviA repressed motility under conditions of low osmolarity . Under conditions of high osmolarity ( 0 . 3 M NaCl ) , the presence or absence of the tviA gene did not alter motility in S . Typhi or S . Typhimurium , suggesting that TviA-mediated repression is relieved under this growth condition [8] ( Figure S2 ) . These observations suggested that the tviA gene is responsible for differences between S . Typhi and S . Typhimurium in expressing the master regulator of flagella expression and that the tviA gene product can be fully incorporated into the regulatory network existing in S . Typhimurium . Furthermore , these data supported the idea that TviA does not affect flagella expression under conditions of high osmolarity ( Figure 2A ) , which are encountered in the intestinal lumen . In contrast , TviA repressed flagella expression under conditions that closely resembled the osmolarity encountered in human tissue . We next wanted to investigate whether TviA-mediated changes in gene transcription altered the amount of flagellin protein produced when S . Typhi strains were grown at an osmolarity encountered in tissue ( i . e . after growth in DMEM tissue culture medium ) ( Figure 2B ) . Expression of the S . Typhi flagellin , FliC ( also known as the S . Typhi Hd antigen ) , was monitored by Western blot ( using anti Hd serum ) . Expression of the heat shock protein GroEL remained constant and was used as a loading control . In the presence of the tviA gene ( i . e . in the S . Typhi wild-type strain or the S . Typhi ΔtviB-vexE mutant ) , a low level of FliC expression was detected when bacteria were grown under conditions mimicking tissue osmolarity ( Figure 2B ) or under conditions of low osmolarity ( Figure S3 ) . Deletion of tviA in S . Typhi ( ΔviaB mutant ) resulted in increased expression of FliC and introducing the cloned tviA gene ( pTVIA1 ) restored FliC expression to wild-type levels . Introduction of the tviA gene into the S . Typhimurium chromosome ( ΔphoN::tviA mutant ) reduced FliC ( also known as the S . Typhimurium H1 or Hi antigen ) protein levels when bacteria were grown in DMEM tissue culture medium ( Figure 2C ) or under conditions of low osmolarity ( Figure S3 ) . Expression of FljB , the H2 flagellin antigen of S . Typhimurium , was not detected by Western blot under conditions used in this study ( data not shown ) . Collectively , these data suggested that TviA reduced the amount of FliC production in S . Typhi and S . Typhimurium under conditions of tissue osmolarity . To further test this idea , we mimicked osmotic conditions encountered in the intestinal lumen or in tissue by suspending green fluorescent protein ( GFP ) -labeled bacteria in medium with high osmolarity or in serum , respectively . After a two-hour incubation , flagella expression was detected on the bacterial surface by flow cytometry . This analysis revealed that flagella were expressed by S . Typhimurium strains under osmotic conditions encountered in intestinal contents , regardless of the presence of tviA ( Figure 3A ) . In contrast , TviA repressed flagellin expression under osmotic conditions encountered in serum , as indicated by a reduction of FliC on the surface of the strain carrying the tviA gene ( i . e . the S . Typhimurium ΔphoN::tviA mutant ) ( Figure 3B ) . Invasion of epithelial cells allows Salmonella to gain access to the lamina propria of the small intestine , a process that is accomplished in as little as two hours [12] . To test , whether tviA can repress flagellin expression within this time frame , the S . Typhimurium ΔphoN mutant and the ΔphoN::tviA mutant were grown under conditions of high osmolarity and subsequently shifted to osmolarity encountered in the tissue . Expression of FliC was determined at different time points by Western blot ( Figure 3C ) . In comparison to the wild-type strain , the tviA gene product reduced the amount of flagellin expression as early as two hours after decreasing the osmolarity of the culture medium . These data were consistent with the hypothesis that TviA does not alter gene expression in the intestinal lumen but rapidly ( within two hours ) represses flagellin expression upon bacterial entry into tissue . To mount responses that are appropriate to the threat , the innate immune system in the intestine needs to distinguish between harmless commensal bacteria that are present in the lumen and pathogenic microbes that invade tissue . One player in this process is the intestinal epithelium , which can discriminate between luminal commensals and invasive pathogens by a functional compartmentalization of Toll-like receptor ( TLR ) 5 expression . TLR5 is a pathogen recognition receptor specific for bacterial flagellin [13] . TLR5 is only expressed on the basolateral surface of the intestinal epithelium [14] , [15] . Human colonic epithelial ( T84 ) cells can be polarized to form a model epithelium that recapitulates the sentinel function of TLR5 in detecting bacterial translocation from the lumen [15] , [16] , [17] . We used this model to investigate whether TviA-mediated repression of flagellin expression in tissue is a mechanism to evade sentinel functions of model epithelia . The expression of CCL20 ( encoding the chemokine MIP-3α ) and CXCL1 ( encoding the chemokine GROα ) in polarized T84 cells was flagellin-dependent , as indicated by an absence of responses elicited by non-flagellated S . Typhi and S . Typhimurium mutants ( Figure 4 , S4 , and Table S1 ) . Furthermore , T84 model epithelia responded to basolateral , but not to apical stimulation with purified flagellin ( Figure 4A ) , which was consistent with a functional compartmentalization of TLR5 expression reported previously [15] . Model epithelia were stimulated basolaterally with S . Typhi strains grown under conditions mimicking tissue osmolarity . The presence of tviA in the S . Typhi wild-type strain and the S . Typhi ΔtviB-vexE mutant resulted in a dramatic reduction in the relative transcript levels of CXCL1 and CCL20 ( Figure 4A and B ) compared to levels elicited by the S . Typhi ΔviaB mutant , which lacked the tviA gene . To determine whether introduction of the tviA gene into S . Typhimurium would confer the ability to evade detection by model epithelia , polarized T84 cells were stimulated basolaterally with S . Typhimurium strains grown under conditions mimicking tissue osmolarity ( Figure 4C ) . The absence of tviA in the S . Typhimurium wild-type strain and the S . Typhimurium ΔphoN mutant resulted in considerable higher mRNA levels of CXCL1 in T84 cells compared to levels elicited by strains in which flagellin expression was repressed ( S . Typhimurium ΔphoN::tviA mutant ) or abrogated ( S . Typhimurium ΔphoN ΔfliC fljB mutant ) . In summary , these data suggested that sentinel functions of the intestinal epithelium could be evaded by a TviA-mediated repression of flagellin expression in tissue . By evading detection through sentinels of the intestinal immune system , TviA-mediated flagellin repression might prevent induction of mucosal barrier functions orchestrated by proinflammatory signals . Since our data pointed to a high degree of similarity between S . Typhi and S . Typhimurium in the mechanisms and consequences of TviA-mediated gene regulation , we reasoned that the relevance of TviA-mediated flagellin repression in vivo could be assessed using animal models of S . Typhimurium infection . The mouse model is not suited for this purpose , because S . Typhimurium rapidly disseminates to the liver and spleen of mice , suggesting that the pathogen can overcome mucosal barrier functions in this host species . In contrast , S . Typhimurium causes a localized gastroenteritis in immunocompetent individuals and is therefore susceptible to mucosal barrier functions encountered in humans . These barrier functions , which are present in humans but absent from mice , are specifically overcome by S . Typhi , as indicated by its ability of to cause typhoid fever . We thus reasoned that the consequences of TviA-mediated flagellin repression should be investigated in an animal , whose mucosal barrier functions , like the ones in humans , are sufficient for preventing systemic dissemination of S . Typhimurium . S . Typhimurium causes a localized enteric infection in chickens , an animal detecting flagellin expression through TLR5 [18] , resulting in the activation of mucosal barrier functions [19] . This host was chosen for our analysis . Groups of four-day-old chickens were infected orally with the S . Typhimurium ΔphoN mutant , the S . Typhimurium ΔphoN::tviA mutant or the S . Typhimurium ΔphoN ΔfliC fljB mutant and bacterial translocation to the spleen was monitored at 8 hours after infection . The presence of the flagellin repressor TviA ( ΔphoN::tviA mutant ) or the absence of flagellin ( ΔphoN ΔfliC fljB mutant ) resulted in markedly increased systemic dissemination of S . Typhimurium compared to that observed with flagellated S . Typhimurium ( ΔphoN mutant ) ( Figure 5 ) . In contrast , no significant differences were detected between numbers of the S . Typhimurium ΔphoN mutant , the S . Typhimurium ΔphoN::tviA mutant or the S . Typhimurium ΔphoN ΔfliC fljB mutant recovered from intestinal contents . Since the flagellin proteins are among the most abundant proteins expressed by S . Typhimurium it was conceivable that TviA increased the growth rate by repressing the flagella regulon . However , the tviA-expressing strain ( ΔphoN::tviA mutant ) and the ΔphoN mutant were recovered in comparable numbers from the spleen of intraperitoneally infected mice 8 h after infection ( Figure S5 ) , indicating that TviA did not alter the growth rate of S . Typhimurium in tissue . Taken together , these data were consistent with the idea that TviA-mediated repression of flagellin expression is a mechanism to overcome mucosal barrier functions , thereby promoting increased bacterial dissemination to the spleen .
The ability to cross epithelial linings is not sufficient for causing systemic bacterial dissemination in immunocompetent individuals , suggesting that additional barrier functions encountered in tissue successfully limit bacterial spread . At least some of these barrier functions are inducible by proinflammatory signals generated during bacterial translocation from the gut [20] . Here we provide support for the idea that evasion of inducible barrier functions by repressing a bacterial PAMP ( i . e . flagellin ) is a mechanism enhancing systemic bacterial dissemination from the intestine . S . Typhimurium expresses flagellin during growth in the intestinal lumen as well as in Payers patch tissue , but flagellin expression ceases once bacteria disseminate to internal organs of mice , such as the spleen [21] , [22] . Our data suggest that TviA-mediated flagellin repression is not operational in the intestinal lumen , but is rapidly initiated once bacteria encounter tissue osmolarity . The presence of TviA might therefore enable S . Typhi to more rapidly repress flagellin expression upon invasion of the intestinal mucosa ( Figure 6 ) compared to S . Typhimurium , which still expresses flagellin in intestinal tissue [21] . Bacterial translocation across the epithelial barrier into the underlying tissue is observed within 2 hours after infection of ligated ileal loops with S . Typhimurium [12] , [23] . TviA markedly reduced flagellin repression within 2 hours of bacterial growth at an osmolarity encountered in tissue . TviA-mediated flagellin repression thus occurred within the time frame required for bacterial translocation across an epithelial barrier in vivo . Similarly , TviA activates expression of the Vi capsular antigen when S . Typhi transits from the intestinal lumen into tissue in a ligated ileal loop model [24] . Expression of flagellin by bacteria arriving in tissue is of consequence , because sentinels monitoring microbial translocation from the gut can detect this PAMP . One of the mechanisms by which the intestinal mucosa distinguishes luminal bacteria from bacteria in tissue can be recapitulated using polarized T84 intestinal epithelial cells , which express TLR5 only on their basolateral surface [15] , [17] . Here we show that TviA-mediated flagellin repression enabled bacteria to evade this sentinel function of epithelial cells . It is possible that other cell types may contribute to detecting flagella in vivo . However , regardless of the mechanism ( s ) by which flagellin stimulates innate immunity in the intestine , our results demonstrate that TviA-mediated flagellin repression resulted in increased bacterial dissemination to the spleen of chickens . The idea that detection of flagella contributes to barrier function is also consistent with the finding that a non-flagellated S . Typhimurium fliM mutant exhibits an enhanced ability to establish systemic infection in chickens compared to the wild-type strain [19] . It may therefore not be a coincidence that S . enterica serotype Gallinarum ( S . Gallinarum ) , the only serotype associated with a severe systemic infection in chickens [25] , does not express flagella . Similarly , tight regulation of flagellin expression is required for virulence of Yersinia enterocolitica in mice [26] . It should be pointed out , however , that evading detection of flagella by the innate immune system , although necessary , might not be sufficient for causing systemic disease . For example , Shigella species cause a localized colitis in humans , despite the fact that these pathogens do not express flagellin . A possible explanation for the lower propensity of Shigella species to cause systemic infection is the absence of a Salmonella T3SS-2 equivalent . T3SS-2 is a Salmonella virulence factor important for macrophage survival [2] , [27] , and its absence in Shigella species may render these pathogens more vulnerable to phagocyte attack . In turn , T3SS-2 may be necessary , but it is not sufficient for systemic dissemination , because S . Typhimurium , which carries this virulence factor , causes a localized infection in immunocompetent individuals . Thus , the ability of S . Typhi to cause systemic disease in humans likely evolved by combining virulence factors conserved among Salmonella serotypes ( e . g . T3SS-2 and others ) with newly acquired virulence traits ( e . g . TviA-mediated flagellin repression and others ) . The picture emerging from these studies is that the presence in S . Typhi of a regulator , TviA , which senses the transition of bacteria from the intestinal lumen into tissue , enables the pathogen to rapidly cease flagellin expression when crossing the epithelial lining , thereby preventing the induction of barrier functions that limit bacterial dissemination ( Figure 6 ) . At the same time , TviA induces expression of the Vi capsular antigen [24] , a virulence factor preventing detection of the pathogen through TLR4 [28] . Collectively , these mechanisms interfere with innate immune surveillance at the mucosal surface [17] , [29] , [30] , [31] , resulting in reduced intestinal inflammation [32] , [33] and contributing to increased dissemination . It should be pointed out that overcoming barrier functions through TviA-mediated regulation is not sufficient for causing typhoid fever , because subsequent to its initial systemic spread , S . Typhi requires additional virulence mechanisms to establish residence in internal organs , persist and , after a two-week incubation period , cause disease .
Bacterial strains and plasmids used in this study are listed in table 1 . Salmonella strains were routinely grown aerobically at 37°C in Luria Bertani ( LB ) broth ( 10 g/l tryptone , 5 g/l yeast extract , 10 g/l NaCl ) or on LB agar plates . To induce optimal expression of TviA , strains were grown overnight in LB , diluted in either Super Optimal Broth ( SOB ) ( 20 g/liter tryptone , 5 g/liter yeast extract , 10 mM NaCl , 2 . 5 mM KCl , 10 mM MgCl2 ) [29] or tryptone yeast extract broth ( 10 g/l tryptone , 5 g/l yeast extract ) and aerobically grown to mid-log phase at 37°C . When appropriate , antibiotics were added at the following concentrations: chloramphenicol 0 . 03 mg/ml , carbenicillin 0 . 1 mg/ml , and kanamycin 0 . 05 mg/ml . Phage P22 HT int-105 was used for transduction as described previously[34] , [35] . To construct strain SW335 , a P22 lysate of strain TH4054 was used to transduce the flhC5456::MudJ mutation into IR715 . SW681 was constructed by transducing the ΔphoN::Kanr mutation of the strain AJB715 into SPN313 . Bacterial RNA was isolated as described previously [8] . Briefly , Salmonella strains were statically grown in 5 ml SOB broth for 2 h . 0 . 8 ml of a 5% phenol solution ( in ethanol ) was added and the bacterial cells collected by centrifugation . The pellet was resuspended in 0 . 4 ml 0 . 1 mg/ml lysozyme , 1 mM ethylenediaminetetraacetic acid ( EDTA ) 10 mM , Tris/Cl pH 8 . 0 and incubated at room temperature for 30 min . Cells were lysed by adding 40 µl 10% sodium dodecyl sulfate ( SDS ) . 0 . 44 ml 1 M sodium acetate as well as 0 . 9 ml hot ( 65°C ) phenol was added to the sample and the emulsion was incubated at 65°C for 6 min , incubated on ice for 10 min and centrifuged at 20 , 000 g for 10 min at 4°C . The upper phase was extracted with 0 . 9 ml chloroform . After centrifugation at 20 , 000 g for 5 min at 4°C , the RNA was precipitated by adding 80 µl 1 mM EDTA 3 M sodium acetate pH 5 . 2 and 1 ml isopropanol . Samples were centrifuged for 30 min at 20 , 000 g at 4°C and the RNA pellet was washed with 1 ml 80% Ethanol . The air-dried RNA was resuspended in RNase-free water and traces of genomic DNA were removed by rigorous DNase treatment according to the recommendation of the manufacturer ( DNA-free DNase treatment , Applied Biosystems ) . Gene expression profiling experiments of the S . Typhimurium strains SW124 and SW125 were conducted identically to experiments described previously [8] . Briefly , RNA was extracted from one bacterial culture grown statically in 5 ml SOB broth until the turbidity reached an optical density of OD600 = 0 . 4−0 . 5 . Microarray hybridization and scanning steps were performed by the UC Davis ArrayCore Microarry facility as described previously [36] with the modifications described in [8] . The TM4 Microarray Software Suite [37] was used for data processing and analysis as described previously [8] . Data from the reference data set ( S . Typhi , [8] ) was averaged and a cluster analysis of the gene expression profile of S . Typhi and S . Typhimurium was performed by the Clustering Affinity Search Technique ( CAST ) algorithm [38] , [39] ( initial threshold parameter of 0 . 85 ) . Genes identified to be regulated by TviA in S . Typhi and S . Typhimurium are listed in supplementary table S1 . Microarray data have been deposited at the Gene Expression Omnibus database under the accession number GSE20321 . Expression of flagellin was determined by Western blot as described previously [8] . In brief , Salmonella strains were grown aerobically for 2 h at 37°C in Dulbecco's Modified Eagle Medium ( DMEM ) ( Invitrogen ) . For time course experiments , Salmonella strains were grown for 16 h in tryptone yeast extract broth containing 0 . 3 M NaCl and diluted in Minimum Essential Medium Eagle ( MEM ) medium ( Invitrogen ) . Culture turbidity ( OD600 ) was measured and bacterial cells were lysed in loading buffer ( 50 mM Tris/HCl , 100 mM dithiothreitol , 2% SDS , 0 . 1% bromophenol blue , 10% glyerol ) . A portion of the lysate corresponding to approximately 5×107 colony forming units ( CFU ) was resolved by SDS-polyacrylamide gel electrophoresis ( PAGE ) [40] . Proteins were transferred onto a polyvinylidene fluoride membrane ( Millipore ) using a semi-dry transfer system ( Bio-Rad laboratories ) . To detect FliC and GroEL expression , rabbit Salmonella H antiserum d ( Difco ) , Salmonella H antiserum i ( Difco ) , and anti-GroEL antiserum ( Sigma ) , respectively , as well as a horse radish peroxidase-conjugated goat anti-rabbit secondary antibody ( Bio-Rad laboratories ) were used . Chemiluminescence ( SuperSignal West Pico Chemiluminescent Substrate , Thermo Scientific ) was detected by a BioSpectrum Imaging System ( UVP ) and images were processed in Photoshop CS2 ( Adobe ) to adjust brightness levels . Salmonella strains were grown overnight in tryptone yeast extract broth , diluted 1∶50 in 5 ml tryptone yeast extract broth and incubated for 3 h at 37°C . To adjust the osmolarity , NaCl was added to the media of the subculture as indicated . β-Galactosidase activity was measured as described previously [8] , [41] . The experiment was performed in triplicate . Strains were grown overnight in LB broth , diluted 1∶50 in fresh LB and incubated at 37°C until log phase . 5×108 CFU were re-suspended in either 0 . 05 ml of mouse serum or in 0 . 05 ml of tryptone yeast extract broth containing 0 . 3 M NaCl and incubated for 2 hours at 37°C . Bacteria were cellected by centrifugation at 6000 g for 5 min at room temperature . Pellets were washed twice in fluorescence activated cell sorting ( FACS ) buffer ( 1% Bovine serum albumin in phosphate buffered saline [PBS] ) and re-suspended in 0 . 1 ml of FACS buffer . Polyclonal rabbit anti-FliC was added and incubated on ice for 30 minutes . A secondary R-PE conjugated goat-anti rabbit ( Jackson ImmunoResearch ) was added and incubated on ice for 30 minutes . Bacteria were fixed in 4% Formalin for 1 hour and analyzed using an LSR II flow cytometer ( Beckton-Dickinson ) . Results were analyzed using FlowJo software ( Treestar ) . The colonic carcinoma cell line T84 was obtained from the American Type Culture Collection ( ATCC , CCL-248 ) . T84 cells were routinely maintained in DMEM-F12 medium containing 1 . 2 g/l sodium bicarbonate , 2 . 5 mM L-glutamine , 15 mM 4- ( 2-hydroxyethyl ) -1-piperazineethanesulfonic acid ( HEPES ) , 0 . 5 mM sodium pyruvate ( Invitrogen ) , and 10% fetal bovine serum ( FBS; Invitrogen ) . To polarize T84 cells , cells were seeded at a density of 1×106 cells per well in the apical compartment of transwell plates ( 12 mm diameter , pore size 0 . 4 µm ) ( Corning ) and incubated for 5 to 10 days until the transepithelial electrical resistance exceeded a value of 1 . 5 kΩcm2 . Media in both compartments was replaced every second day . Salmonella strains were grown over night at 37°C in LB , diluted 1∶50 in yeast extract broth or MEM medium ( Invitrogen ) and incubated for 2 h 30 min at 37°C with aeration . T84 cells were activated by adding 2×106 CFU into the basolateral compartment containing 1 ml of media . Purified Salmonella flagellin ( InvivoGen ) was added into the indicated compartment at a concentration of 1 µg/ml . After 3 h , eukaryotic RNA was isolated as described previously [11] using TRI reagent ( Molecular Research Center ) In brief , cells were lysed in 0 . 5 ml TRI reagent and this homogenate extracted with 0 . 1 ml chloroform ( Sigma ) . The suspension was centrifuged at 12 , 000 g for 15 min . Nucleic acids were precipitated from the aqueous phase by adding 0 . 25 ml isopropanol ( Sigma ) and by centrifugation at 12 , 000 g for 8 min . The RNA pellet was washed with 75% Ethanol , air-dried and resuspended in water . Traces of DNA were removed by DNase treatment according to the recommendation of the manufacturer ( DNA-free DNase treatment , Applied Biosystems ) . Real-time quantitative ( q ) reverse transcriptase ( RT ) -polymerase chain reaction ( PCR ) was performed as described previously [11] . 1 µg of DNase treated bacterial or eukaryotic RNA served as a template for RT-PCR in a 50 µl volume . Random hexamer dependent amplification was performed according to the recommendations of the manufacturer ( TaqMan reverse transcription reagents; Applied Biosystems ) . SYBR Green ( Applied Biosystems ) based real-time PCR was performed in an 11 µl volume employing 4 µl of cDNA as a template . Primers are listed in table 2 and were added at a final concentration of 250 nM . Primers used to detect expression of bacterial genes were designed to amplify targets from both Salmonella serotypes with equal efficiency . Data was acquired by a GeneAmp 7900 HT Sequence Detection System ( Applied Biosystems ) and analyzed using the comparative Ct method ( Applied Biosystems ) . Bacterial gene transcription in each sample was normalized to the respective levels of guanylate kinase mRNA , encoded by the gmk gene . Eukaryotic gene expression was normalized to the respective levels of GAPDH mRNA . All procedures described in this study were conducted as described previously [42] . Briefly , specific pathogen free eggs were obtained from Charles River ( North Franklin , CT ) . Eggs were kept in an egg incubator at 38°C and a humidity of 58–65% for 21 days and were periodically rolled for the first 18 days . Chickens were housed in a poultry brooder ( Alternative Design Manufacturing , Siloam Springs , AR ) at a temperature of 32°C to 35°C . Tap water and irradiated lab chick diet ( Harlan Teklad , Madison , WI ) was provided ad libitum . S . Typhimurium strains were grown aerobically at 42°C for 16 h in LB broth . Fifteen 4-day-old , unsexed White Leghorn chicks were orally inoculated in groups of five with either 1×109 CFU of the S . Typhimurium strains AJB715 , SW474 , or SW681 in 0 . 1 ml LB broth . Animals were euthanized by asphyxiation with CO2 8 h after inoculation . The spleen and a sample of the cecal content were homogenized in sterile PBS and serial ten-fold dilutions spread on LB agar plates containing the appropriate antibiotics . C57BL/6 mice were obtained from The Jackson Laboratory . Animals were housed under specific-pathogen-free conditions and provided with water and food ad libitum . S . Typhimurium strains were grown aerobically for 16 h at 37°C . Groups of 4 female mice ( 10 to 11 weeks of age ) were injected intraperitoneally with 1×106 CFU of the S . Typhimurium strains IR715 , AJB715 , or SW474 suspended in PBS . 8 h after infection , animals were euthanized and the spleen collected . Serial 10-fold dilutions of the splenic homogenate were spread on LB agar plates containing nalidixic acid . For the statistical analysis of ratios ( i . e . increases in gene expression ) , values were transformed logarithmically for further statistical analysis . Data presented in bar graphs are geometric means +/− standard error . A parametric test ( Student's t-test ) was used to determine whether differences between treatment groups were statistically significant ( P<0 . 05 ) . For data from tissue culture experiments and gene expression analysis , paired statistical analysis was used . All animal experiments were performed according to Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) guidelines . Experimental procedures with chickens were approved by the Texas A&M University Institutional Animal Care and Use Committee ( IACUC ) . All experimental procedures with mice were approved by the UC Davis IACUC .
|
Some bacterial species contain pathogenic strains that are closely related genetically , but cause diseases that differ dramatically in their clinical presentation . One such species is Salmonella enterica , which contains non-typhoidal serotypes associated with a localized gastroenteritis and serotype Typhi ( S . Typhi ) , the causative agent of a severe systemic infection termed typhoid fever . Conventional wisdom holds , that the ability of S . Typhi to overcome mucosal barriers and spread systemically in immunocompetent individuals evolved through acquisition of new virulence factors , which are absent from non-typhoidal Salmonella serotypes . Here , we demonstrate that acquisition of a regulatory gene , tviA , by S . Typhi alters expression of existing virulence factors ( the flagellar regulon ) such that molecular structures that are detected by the host innate immune are repressed after entering tissue . We propose that this mechanism contributes to innate immune evasion by S . Typhi , thereby promoting systemic dissemination .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"microbiology/innate",
"immunity",
"infectious",
"diseases/bacterial",
"infections",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis",
"microbiology/medical",
"microbiology",
"infectious",
"diseases/gastrointestinal",
"infections"
] |
2010
|
A Rapid Change in Virulence Gene Expression during the Transition from the Intestinal Lumen into Tissue Promotes Systemic Dissemination of Salmonella
|
Tissue-resident memory T cells are required for establishing protective immunity against a variety of different pathogens , although the mechanisms mediating protection by CD4+ resident memory T cells are still being defined . In this study we addressed this issue with a population of protective skin-resident , IFNγ-producing CD4+ memory T cells generated following Leishmania major infection . We previously found that resident memory T cells recruit circulating effector T cells to enhance immunity . Here we show that resident memory CD4+ T cells mediate the delayed-hypersensitivity response observed in immune mice and provide protection without circulating T cells . This protection occurs rapidly after challenge , and requires the recruitment and activation of inflammatory monocytes , which limit parasites by production of both reactive oxygen species and nitric oxide . Overall , these data highlight a novel role for tissue-resident memory cells in recruiting and activating inflammatory monocytes , and underscore the central role that skin-resident T cells play in immunity to cutaneous leishmaniasis .
Tissue-resident memory T cells ( TRM ) are critical mediators of immunity against a number of different infections in a variety of different tissues [1–11] . Because they are typically located at barrier surfaces and therefore occupy the initial sites of infection , TRM cells are poised to provide rapid protection . CD8+ TRM cells are the best defined tissue-resident T cells , and mediate protection through direct cytotoxicity [12–14] , production of cytokines [1 , 15] , maturation of local innate cells [6] , triggering of tissue-wide antiviral signaling [16] , and/or the recruitment of additional lymphocytes to the site of infection [15] . CD4+ TRM cells remain relatively uncharacterized , although they have been described in the lung , vaginal mucosa , and skin [3–5 , 17] . We recently demonstrated that skin-resident CD4+ T cells play a critical role in immunity to cutaneous leishmaniasis [18] , however the various mechanisms by which CD4+ TRM cells mediate protection in the skin remain ill-defined . Human cutaneous leishmaniasis encompasses a spectrum of diseases caused by the intracellular protozoan parasites . Murine models that mimic aspects of the human disease have proven invaluable for understanding the mechanisms mediating susceptibility and resistance [19] . For example , similar to some forms of human cutaneous leishmaniasis , C57BL/6 mice infected with Leishmania major develop lesions that heal over several weeks , and once resolved the mice exhibit immunity to reinfection [19] . Studies in this model have shown that in a primary leishmania infection , innate cells including neutrophils , monocytes , and dendritic cells are rapidly recruited to the site of challenge [20–23] . These cells have the potential to restrict parasite infection [21 , 24–26] , but they can also be co-opted by the parasites to evade immune detection or suppress the immune response [20 , 27 , 28] . Conversely , in a secondary infection , the recruitment of pre-existing circulating effector CD4+ Th1 cells leads to the rapid control of the parasites [29 , 30] , and CD4+ TRM cells contribute by promoting the recruitment of these effector T cells to the site of infection [18] . However , given their location at the site of a challenge infection and their rapid production of IFNγ , it might be expected that CD4+ TRM cells may also provide some level of rapid protection that is independent of additional T cell recruitment from the blood . Here we show that CD4+ TRM cells mediate control of the parasite burden within the first three days of infection , which correlates with a strong delayed-type hypersensitivity ( DTH ) response , the hallmark of immunity in murine and human leishmaniasis . While IFNγ produced by TRM cells might be expected to activate resident macrophages in the skin and limit the parasite burden , surprisingly we found that protection by CD4+ TRM cells required the recruitment of inflammatory monocytes that subsequently controlled the parasites by the induction of both reactive oxygen species ( ROS ) and inducible nitric oxide synthase ( iNOS ) . Importantly , we found that TRM cells provided protection independently of circulating CD4+ T cells , emphasizing the importance of generating TRM cells for optimal immunity to leishmaniasis .
In experimental models of cutaneous leishmaniasis , protection to a challenge infection is often assessed after several weeks , when a large difference in parasite number is evident between naive and immune mice . This approach also allows for the assessment of protection mediated not only by circulating effector T cells , but also by central memory T cells that are delayed in their protective response [30] . However , the identification of TRM cells and their occupation of the skin led us to hypothesize that they might contribute to immune protection very early after challenge . To test this , we challenged naive and leishmania-immune mice in the ear with L . major , and assessed the immune response during the first 72 hrs of infection . For these studies , immune mice were infected with L . major in the contralateral ear at least 12 weeks earlier , and had resolved their primary lesion . One of the hallmarks of immunity to leishmaniasis is the presence of a DTH response , and a positive reaction indicates that an individual has generated a type 1 immune response . As expected , immune mice developed a DTH response , represented by an increase in ear swelling within 24–72 hrs after challenge , while naive mice did not ( Fig 1A ) . In order to evaluate whether the presence of this DTH reaction was associated with control of the challenge inoculum , we assessed the parasite burden by performing three different assays: limiting dilution , qPCR for parasite ribosomal RNA , and analysis of the frequency of cells infected with dsRed expressing parasites by flow cytometry . We found that the number of parasites was consistently decreased 2–4 fold in immune mice at 72 hrs , as measured by limiting dilution and qPCR ( Fig 1B and 1C ) , and that the frequency of infected cells was significantly decreased by flow cytometry ( Fig 1D ) . These results demonstrate that as early as 72 hrs after challenge , mice that have resolved a previous L . major infection can mount an immune response that is effective at controlling the parasites . To determine if the DTH in leishmaniasis was dependent on either CD4+ or CD8+ T cells , we individually depleted each subset in immune mice before challenge with L . major , and then monitored the DTH response and parasite burden over 72 hrs ( Fig 1E ) . αCD4 treatment , which depletes both circulating and tissue-resident CD4+ cells in our hands ( S1 Fig ) , completely ablated the DTH response ( Fig 1F ) , while effective CD8 depletion ( S1 Fig ) did not , suggesting that CD4+ cells are the critical drivers of this early inflammation . Importantly , CD4+ cells , but not CD8+ T cells , were also required for the decrease in parasites at 72 hrs ( Fig 1G ) . We next wanted to test if TRM cells were mediating the early control of the parasites . To do so , we grafted naive and immune skin side-by-side onto the flanks of naive recipient mice , challenged each graft , and measured the parasite burden three days later ( Fig 1H ) . As the graft recipients contain only naïve T cells , this approach enabled us to specifically assess the protection mediated by TRM cells , which we have previously shown to remain in the grafted tissue [18] . In all cases , the immune grafts had significantly fewer parasites than their naive counterparts at 72 hrs ( Fig 1I ) . Taken together , these results indicate that CD4+ TRM cells mediate parasite protection in immune skin at 72 hrs in a process that is independent of circulating CD4+ and CD8+ T cells . To gain further insight into how this rapid protection is mediated , we analyzed the cells recruited to the skin of naive and immune mice 72 hrs after challenge . We compared the numbers of CD90 . 2+ T cells , Ly6G+ neutrophils , Ly6C+ inflammatory monocytes , MerTK+ CD64+ macrophages , and CD11c+ MHCII+ dendritic cells in naive and immune skin 72 hrs after infection ( Fig 2A ) . As expected , we observed increased T cell recruitment to immune skin consistent with previous results [18 , 29] . However , a majority of the recruited cells were myeloid lineage cells , specifically inflammatory monocytes ( Fig 2B and 2C ) . We analyzed the activation status of these monocytes and found that they expressed high levels of MHCII , ROS , and iNOS . Further , both MHCII and iNOS expression were significantly increased in the monocytes recruited to immune skin compared with those recruited to naive skin ( Fig 2D ) . Finally , using fluorescent parasites , we found that greater than 70% of the infected cells in the skin of immune mice were inflammatory monocytes ( Fig 2E ) . Notably , these infected cells contained fewer parasites per cell when compared with monocytes in naive skin , as demonstrated by the lower MFI of dsRed ( Fig 2F and 2G ) , and when counted in cytospins ( Fig 2H ) . These data show that monocytes are highly recruited to immune skin where they are more likely to be infected than other cell types , have a more activated phenotype , and contain fewer parasites per infected cell . These results suggest that inflammatory monocytes recruited by TRM cells might be better able to kill parasites , and therefore we next investigated whether they were required for parasite control and if so how they mediated protection . To assess the role of inflammatory monocytes in the early protection of immune mice , we used a pair of depleting antibodies that target either Ly6G+ and Ly6C+ cells ( and thus deplete both neutrophils and monocytes ) or Ly6G+ cells alone ( and thus deplete only neutrophils ) ( Fig 3A , S2 Fig ) . Depletion of neutrophils alone had no effect on the DTH response in immune mice , but depletion of both neutrophils and monocytes dramatically reduced the early inflammatory response ( Fig 3B ) . Importantly , the protection observed in immune mice was also completely ablated by the depletion of monocytes and neutrophils , while depleting neutrophils alone did not have a significant effect ( Fig 3C ) . Because activated CD4+ cells can also express Ly6C , we confirmed that RB6-8C5 treatment did not reduce the frequency of Ly6C+ CD4+ T cells in the spleen or ear after challenge ( S3 Fig ) , or the frequency of leishmania-specific IFNγ+ skin-resident T cells ( S4 Fig ) , which are intermediate for Ly6C expression ( S5 Fig ) . Taken together , these data suggest that inflammatory monocytes are the critical mediators of early protection . To specifically address whether inflammatory monocyte recruitment is critical to early protection against L . major , we assessed the response to challenge in CCR2-/- mice , which contain monocytes that lack the ability to respond to CCL2 and CCL7 chemokine signaling and therefore cannot be efficiently recruited to sites of inflammation [31 , 32] . To do so , we grafted naive and immune skin from WT mice onto the flanks of naive WT or CCR2-/- recipients , challenged with L . major , and measured the parasite burden 72 hrs later ( Fig 3D ) . As previously observed , immune skin had significantly fewer parasites compared to naive skin in WT recipient mice ( Fig 3E ) . In contrast , the reduction of parasites in immune skin was lost in CCR2-/- recipient mice ( Fig 3E ) , and correlated with a loss of activated monocytes in the skin ( Fig 3F ) . Together , these results demonstrate that it is recruited CCR2+ monocytes , rather than resident myeloid cells , that are required for protection . To further confirm that inflammatory monocytes were necessary for early protection , and that this protection could be conferred in the absence of circulating T cells , we grafted WT naïve and immune skin onto the flanks of RAG-/- recipient mice that lack T and B lymphocytes . Additionally , we treated some of the mice with αGR1 clone RB6-8C5 to deplete inflammatory monocytes and neutrophils as described above . We challenged each graft with L . major and measured the parasite burden 72 hrs later ( Fig 3G ) . As expected , immune grafts on RAG-/- mice contained significantly fewer parasites , demonstrating that the protection observed at 72 hrs was independent of circulating lymphocytes ( Fig 3H ) . Protection in immune skin was lost in mice treated with RB6-8C5 ( Fig 3H ) . When we quantified the number of Ly6C+MHCII+ cells in each graft , we found a strong correlation with the level of protection ( Fig 3I ) . Taken together , these data further implicate inflammatory monocytes as the critical cell type required for early protection . To gain further insight into the mechanism by which the inflammatory monocytes might control the parasites , we performed skin graft experiments in which we grafted naive and immune skin onto the flank of 1 ) WT naive mice , 2 ) Phox-/- mice in which monocytes lack the ability to produce ROS , or 3 ) iNOS-/- mice that have deficient NO production ( Fig 3J ) . Grafts were then challenged with L . major , and the parasite burden measured at 72 hrs . We found that the protection associated with immune skin was lost in both the Phox-/- and iNOS-/- mice ( Fig 3K ) , suggesting that both ROS and NO from inflammatory monocytes are required for this early protection . Although TRM cell-mediated recruitment of inflammatory monocytes was sufficient to reduce the parasite burden at 72 hrs , we predicted that the presence of circulating leishmania-specific T cells might further enhance immunity , as they are also recruited early after challenge [18 , 29] . To examine the contribution of circulating T cells , we pretreated immune mice with either FTY-720 , which prevents egress of T cells from tissues , or αCXCR3 , which we previously demonstrated blocks the ability of TRM cells to recruit effector T cells from circulation [18] ( Fig 4A ) . Despite the expected decrease in the number T cells recruited to the challenge site ( S6 Fig ) , neither treatment affected the DTH response ( Fig 4B ) , the decrease in parasite burden ( Fig 4C ) , or the recruitment of monocytes ( Fig 4D ) . Unexpectedly , this result demonstrates that DTH and early parasite control are not enhanced by T cells from circulation , and implies that CD4+ TRM cells are solely responsible for mediating these responses . To test if circulating T cells would provide protection in the absence of TRM cells , we utilized a parabiotic model in which the circulations of naive and immune mice were surgically joined , allowing circulating T cells to equilibrate between the two animals , while TRM cells remained exclusively in the immune partner ( Fig 4E , S7 Fig ) . Each parabiont was then challenged with L . major in the ear , and DTH and parasite number were measured 72 hrs later . As expected , immune parabionts had the same DTH response , monocyte recruitment , and parasite numbers as control immune mice ( Fig 4F–4H ) . In contrast , naive parabionts , despite having a full complement of circulating memory T cells , did not exhibit a DTH response , had defective monocyte recruitment , and lost the early protection observed in immune mice ( Fig 4F–4H ) . These results show that circulating leishmania-specific T cells by themselves are unable to provide any protection at this early time point , further demonstrating that TRM cells are the critical subset for this rapid protection . In contrast to the parasite control observed at 72 hrs in immune mice , we previously found that when protection was assessed two weeks after challenge with 2 x 106 parasites , optimal parasite control depended upon both CD4+ TRM cells and circulating effector T cells [18] . These results , in combination with our current findings , suggest that while TRM cells may initially reduce the parasite number , the long-term consequences are limited in the absence of additional circulating T cells . However , since parasite dose can significantly influence what is required for protection , and the number of infective parasites transmitted by the sand fly is thought to be much lower than 2 x 106 parasites [33] , we next tested whether TRM cells might provide protection greater than 72 hrs after challenge if fewer parasites were present in the challenge inoculum . First , we challenged naive and immune mice with 103 parasites in the ear , measured the DTH , and assessed the composition of cells recruited to the challenge site at 72 hrs . Similar to our results with high dose challenge , immune mice had an increased DTH response ( Fig 5A ) , there was a large population of inflammatory monocytes infiltrating the lesions ( Fig 5B ) , and the monocytes had a more activated phenotype ( Fig 5C ) , though the magnitude of the overall response was lower . To test if TRM cells could provide protection more than 72 hrs after challenge without circulating T cells , we challenged intact naive mice , immune mice , naive grafts , and immune grafts with 103 L . major , and measured parasite burdens 4 weeks later ( Fig 5D ) . As expected , intact immune mice were better protected than naive mice against low dose L . major challenge ( Fig 5E ) . However , immune skin grafts also showed significantly better protection than their naive counterparts , despite the absence of circulating leishmania-specific effector T cells ( Fig 5E ) . These results demonstrate that TRM cells do not require previously activated circulating T cells to provide protection as long as 4 weeks after low dose challenge . However , naive leishmania-specific T cells would be expanded during the 4 weeks of infection , and could contribute to the protection we measured . Therefore , to test if TRM cells could provide protection without any circulating lymphocytes , we grafted naive and immune skin onto WT and RAG-/- recipient mice , challenged with 103 parasites , and measured the parasite burden 4 weeks later ( Fig 5D ) . Surprisingly , we found that the immune grafts showed significantly better control of the parasites in both WT and RAG-/- mice ( Fig 5F ) . While circulating effector T cells may have potential to contribute to long-term immunity , these results indicate that TRM cells and innate cells alone are sufficient to provide a significant level of protection .
We recently reported that TRM cells provide optimal immunity against L . major infection by recruiting circulating leishmania-specific effector T cells to the site of infection [18] . We now identify an additional , novel function for leishmania-specific TRM cells: to rapidly recruit and activate inflammatory monocytes at the site of infection , resulting in a significant reduction in the initial parasite burden . Further , we show that when the challenge inoculum is at a physiologically relevant dose , CD4+ TRM and inflammatory monocytes exhibit significant control of the parasites , even when circulating leishmania-specific effector T cells are not present . Together , these results demonstrate that in addition to facilitating the recruitment of circulating effector T cells , CD4+ TRM cells play a primary role in controlling parasites immediately after challenge , which not only indicates the importance of generating CD4+ TRM cells in a vaccine , but also expands our understanding of the functions of CD4+ TRM cells . Our experiments have identified CD4+ TRM cells as the critical cell subset required for both the DTH response and the immediate control of leishmania infection . The response is antigen specific as it is not induced by PBS injection , and is likely initiated via local antigen presentation [34] . The identification of TRM cells as required for DTH responses was unexpected , since the prevailing view was that circulating effector T cells mediated DTH . However , our results are similar to those that have been described in studies of contact-hypersensitivity , where TRM cells mediated the inflammatory response independent of circulating T cells [35] . Since DTH responses can be elicited at sites distal to the initial site of infection , these results confirm that CD4+ TRM are distributed and can function throughout the skin . Thus , our results extend those of others which have focused on the functions of CD4+ TRM cells at the site of infection [3 , 17] . Surprisingly , neither CD8+ T cells nor circulating effector T cells are required for the DTH response and the early control of the parasites , implicating CD4+ TRM as the mediators of the initial inflammatory responses . We found that inflammatory monocytes are rapidly recruited to the lesion site by TRM cells , and are responsible for the observed protection . Inflammatory monocytes are important mediators of protection against many viral [36 , 37] bacterial [38–40] , fungal [41 , 42] , and parasitic infections [24 , 43 , 44] , and thus this mechanism of protection has the potential to influence a number of different immune responses . Inflammatory monocytes are known to be activated by memory T cells [45] , and restrict leishmania infection in certain contexts [21 , 24 , 44] , but their role in secondary leishmania infection and interaction with TRM cells has not previously been appreciated . On the other hand , neutrophils did not appear to be required for early protection or the recruitment of inflammatory monocytes . This is in contrast to a primary infection where neutrophils may contribute to the recruitment of dendritic cells [46] . This difference is most likely due to the presence of TRM cells , which are sufficient to mediate phagocyte recruitment in secondary challenge . Indeed , CCL2 and CCL7 transcripts are both increased 12 hrs after challenge in immune mice [18] , and CCR2 signaling is required for the recruitment of inflammatory monocytes and subsequent protection . ROS and NO from myeloid lineage cells have both been shown to have roles in controlling leishmania infections , though results vary based on the site of infection , the species of parasite involved , and whether the studies are done in mice or humans [21 , 47–51] . Nonetheless , it is clear in our model that both ROS and NO are required for full protection early after challenge of immune mice , and neither is sufficient alone . Thus , while most studies in mice have emphasized the central role of NO , it has become clear that ROS can contribute to protection not only in humans but also in the mouse . For example , following L . major infection Phox-/- mice can develop chronic lesions long after presumed cure [47] . Understanding why ROS is required under certain conditions for control of leishmanial infections in mice is still not well understood , although we would speculate that at the early time point the levels of NO induced may be insufficient for parasite control , and ROS are required to boost the killing by the inflammatory monocytes . Currently there is no human vaccine for leishmaniasis , which has been partially attributed to the inability to maintain sufficient circulating effector T cells following immunization [52–54] . Thus , it is the presence of low numbers of parasites in immune mice that are believed to maintain maximal levels of responsive effector cells [29 , 30 , 52 , 54] . However , our studies indicate that TRM cells can mediate protection alone , suggesting that these T cells should be targeted for a vaccine . Importantly , we found that they can survive in the absence of persistent parasites [18] , similar to central memory T cells [30] , and thus if generated in a vaccine may be maintained long-term . Thus , defining the requirements for the generation and maintenance of TRM cells , as well as developing vaccination strategies that induce TRM , are the next important steps in developing a vaccine for leishmaniasis . As our understanding of tissue resident T cells grows , more functions have been attributed to TRM cells . CD8+ TRM can be directly cytotoxic [12 , 13] , and IFNγ from TRM cells has been shown to drive recruitment of circulating T cells [15 , 18] . Transcriptional analyses have identified a core set of changes induced by TRM activation that induce a tissue state of pathogen alert , capable of protecting against viral challenge non-specifically , [6 , 16] . However , this is the first demonstration to our knowledge of TRM cells orchestrating the innate response and classic DTH responses by recruiting inflammatory monocytes to the site of infection . This protective mechanism has the potential to be relevant for a number of different intracellular infections , as DTH responses are the hallmark of immunity against many infections and inflammatory monocytes are potent killers of many pathogens [55] . Although circulating effector T cells are undoubtedly beneficial , in certain contexts the rapid response provided by a combination of CD4+TRM cells and inflammatory monocytes that lessen the initial pathogen burden may be critical in limiting the magnitude of the disease .
This study was conducted according to the principles specified in the Declaration of Helsinki and under local ethical guidelines ( University of Pennsylvania Institutional Review Board ) . This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee , University of Pennsylvania Animal Welfare Assurance Number 805186 . C57BL/6 mice were purchased from the National Cancer Institute ( Fredericksburg , MD ) . CCR2-/- ( B6 . 129S4-Ccr2tm1Ifc/J ) , Phox-/- ( B6 . 129S-Cybbtm1Din/J ) , iNOS-/- ( B6 . 129P2-Nos2tm1Lau/J ) , and RAG-/- ( B6 . 129S7-Rag1tm1Mom/J ) mice were purchased from The Jackson Laboratory . All mice were maintained in a specific pathogen-free environment at the University of Pennsylvania Animal Care Facility . L . major ( Friedlin ) or dsRed+ L . major ( Friedlin ) parasites were grown in complete Schneider's insect medium ( GIBCO ) supplemented with 20% heat-inactivated FBS , 2mM glutamine , 100 U/ml penicillin , 100 μg/ml streptomycin , and 50 μg/mL G418 sulfate ( Cellgro ) ( CSM ) . Metacyclic enriched promastigotes were used for infection [56] . Mice were infected with 2 x 106 L . major or 103 L . major intradermally in the ear or flank skin as noted . For flow cytometry analysis αCD45 APC-eF780 , αCD45 . 2 FITC , αCD45 . 1PE-Cy7 , αCD90 . 2 BV605 , αCD11b BV650 , αCD4 PE TexasRed , αCD8b PerCp/Cy5 . 5 , αLy6C AF700 , αLy6G PacBlue , αMerTK APC , αCD64 PE-Cy7 , αCD11c FITC , αMHCII APC , αAF488 iNOS were incubated with single cell suspensions 30 minutes at 4°C and read on LSR Fortessa . For ROS stain , 2ng/mL dihydrorhodamine 123 ( DHR , Cayman Chemical ) was added directly ex vivo , then incubated 30 minutes at 37°C for 30 minutes . For in vivo blockade/depletion 250 μg of αCD4 ( GK1 . 5 ) , αCD8 ( 53–6 . 72 ) , αCXCR3 ( CXCR3-173 ) , 500 μg of αGR1 ( RB6-8C5 ) , αLy6G ( 1A8 ) ( BioXcell ) , or 1 mg/kg FTY-720 ( Cayman Chemical ) were given i . p . one day before challenge . For ear preparation , dorsal and ventral layers of the ear were separated and incubated in RPMI ( Gibco ) with 250 μg/mL Liberase TL ( Roche ) for 90 minutes at 37°C in 5% CO2 . Skin was then dissociated using a 40 μm cell strainer ( BD Pharmingen ) and resuspended in complete RPMI media ( cRPMI ) containing 10% FBS , 100 U/ml penicillin , 100 μg/ml streptomycin , and 55 μM 2-Mercaptoethanol . For flank skin preparation , a section of skin was harvested from the flank following hair removal with an electric trimmer equipped with a two-hole precision blade ( Wahl ) . Skin sections were then minced with a sterile scalpel blade into ~2mm sections , and incubated in RPMI containing 1 mg each of type III and type IV collagenase ( Worthington ) for 120 minutes with vortexing every 30 minutes . The resulting solution was passed through a 40 μm cell strainer and resuspended in cRPMI . Bone marrow derived dendritic cells for restimulations were generated by culturing C57BL/6 bone marrow in GM-CSF supplemented cRPMI for 7–11 days . BMDCs were then harvested and infected 5–8 hours with stationary phase L . major at a ratio of 10:1 in the presence of 1 μg/ml CpG and LPS . Infected BMDCs were incubated at a ratio of 1:5 with 106 skin cells in 24 well plates for 12–16 hours . Cells were incubated for the last 4 hours with 5 μg/ml BFA ( eBioscience ) , stained for IFNγ , and analyzed by flow cytometry . Parasite burden from ear and flank skin was calculated by serial 2-fold dilution in 96-well plates of CSM and incubated at 26°C . The number of viable parasites was calculated from the highest dilution at which parasites were observed 7 days into culture . For qPCR , single cell suspensions from infected tissue were diluted in RLT lysis buffer , then RNA was isolated using the RNeasy Plus kit ( Qiagen ) . RNA was converted to cDNA using the High Capacity RNA to cDNA kit ( Applied Biosciences ) , then the Power SYBR green PCR mater mix ( Applied Biosciences ) was used to quantify parasite ribosomal ssRNA on the ViiA7 qPCR machine ( Applied Biosciences ) with primers F: 5'-TACTGGGGCGTCAGAG-3' and R: 5'-GGGTGTCATCGTTTGC-3' . Cytospins were prepared at 1000 RPM ( Shandon Cytospin3 ) and imaged by light microscopy at 40X magnification ( Nikon E600 ) . Skin grafts were performed as previously described [18] . Briefly , donor skin was prepared under sterile conditions from naive and immune mouse flank skin by shaving , depilating , cleaning with chlorhexidine ( Vetoquinol ) , then excising the skin using sterile 8mm biopsy punches ( Miltex ) . Grafts were placed onto a fresh graft bed prepared by excising skin using a 6mm biopsy punch . All mice were anesthetized , received analgesics , and were monitored post-operatively as previously described . In challenge experiments , graft skin was injected intradermally with 2 x 106 metacyclic L . major 14–20 days after grafting . Congenically disparate ( CD45 . 1+ naive and CD45 . 2+ immune ) mice were cohoused 2 weeks prior to surgery . After induction of anesthesia with isoflorane , each received 0 . 1mg/kg buprenorphine subcutaneously as preemptive analgesia . The surgical site was shaved and aseptically prepared with chlorhexidine scrub . A longitudinal skin incision was made on the mirroring side in each mouse starting at 0 . 5 cm above the elbow and ending 0 . 5 cm below the knee joint . The left elbow and knee of one animal were attached to the right elbow and knee of the other with a 3–0 ethilon suture ( Ethicon ) around each joint beneath the skin in a manner loose enough to not disrupt circulation to the distal limb . The dorsal and ventral skin edges created by the flank incision from one mouse were sutured to the respective skin edges of the second mouse using a continuous absorbable 5–0 vicryl suture patter ( Ethicon ) . Suture glue ( Abbott laboratories ) was used to approximate skin edges . 0 . 5 ml of 0 . 9% NaCl was administered subcutaneously to each mouse to prevent dehydration in the immediate post-operative recovery period , and mice were monitored twice daily for the first 48 hrs post-operatively , then observed daily for signs of surgical site complications , pain , or discomfort . In challenge experiments , ears were infected intradermally with 2 x 106 metacyclic L . major 14–20 days after surgery . Statistical analysis was performed with the Student's t-test ( paired or unpaired where applicable ) , ANOVA , or 2-way ANOVA in Prism software ( GraphPad ) .
|
Cutaneous leishmaniasis is a neglected tropical disease , causing significant worldwide morbidity . There is no vaccine for this infection , in part because of our limited understanding of the memory T cells that might contribute to immunity . We previously discovered that a population of skin-resident memory CD4+ T cells that develop in immune mice enhances the protective immune response against leishmania parasites . Here we show that these skin-resident T cells mediate protection within the first three days of infection . This protection was dependent upon the recruitment of inflammatory monocytes to the challenge site , which reduced the parasite burden in a nitric oxide and reactive oxygen species dependent manner . A series of experiments including blockade of cell recruitment from the blood to the lesions , skin grafts , and parabiosis demonstrated that circulating effector T cells do not contribute to this early protection . Together , these results emphasize that skin-resident CD4+ T cells play a primary role in controlling parasites immediately after challenge , which not only indicates the importance of generating these cells in a vaccine , but also expands our understanding of the functions of skin-resident CD4+ T cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"plastic",
"surgery",
"and",
"reconstructive",
"techniques",
"ears",
"immunology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"skin",
"grafting",
"protozoans",
"signs",
"and",
"symptoms",
"leishmania",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"t",
"cells",
"head",
"immune",
"response",
"diagnostic",
"medicine",
"cell",
"biology",
"monocytes",
"leishmania",
"major",
"anatomy",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2017
|
Skin-resident CD4+ T cells protect against Leishmania major by recruiting and activating inflammatory monocytes
|
Meiotic chromosome segregation relies on homologous chromosomes being linked by at least one crossover , the obligate crossover . Homolog pairing , synapsis and meiosis specific DNA repair mechanisms are required for crossovers but how they are coordinated to promote the obligate crossover is not well understood . PCH-2 is a highly conserved meiotic AAA+-ATPase that has been assigned a variety of functions; whether these functions reflect its conserved role has been difficult to determine . We show that PCH-2 restrains pairing , synapsis and recombination in C . elegans . Loss of pch-2 results in the acceleration of synapsis and homolog-dependent meiotic DNA repair , producing a subtle increase in meiotic defects , and suppresses pairing , synapsis and recombination defects in some mutant backgrounds . Some defects in pch-2 mutants can be suppressed by incubation at lower temperature and these defects increase in frequency in wildtype worms grown at higher temperature , suggesting that PCH-2 introduces a kinetic barrier to the formation of intermediates that support pairing , synapsis or crossover recombination . We hypothesize that this kinetic barrier contributes to quality control during meiotic prophase . Consistent with this possibility , defects in pch-2 mutants become more severe when another quality control mechanism , germline apoptosis , is abrogated or meiotic DNA repair is mildly disrupted . PCH-2 is expressed in germline nuclei immediately preceding the onset of stable homolog pairing and synapsis . Once chromosomes are synapsed , PCH-2 localizes to the SC and is removed in late pachytene , prior to SC disassembly , correlating with when homolog-dependent DNA repair mechanisms predominate in the germline . Indeed , loss of pch-2 results in premature loss of homolog access . Altogether , our data indicate that PCH-2 coordinates pairing , synapsis and recombination to promote crossover assurance . Specifically , we propose that the conserved function of PCH-2 is to destabilize pairing and/or recombination intermediates to slow their progression and ensure their fidelity during meiotic prophase .
During sexual reproduction , meiotic chromosome segregation generates haploid gametes so that fertilization restores diploidy to the resulting embryo . Meiosis halves the DNA complement by segregating homologous chromosomes in one division and follows that reductional division with an equational division that resembles mitosis , in which sister chromatids are partitioned . Proper segregation of homologous chromosomes during meiosis I requires the formation of a linkage , or chiasma , between homologs . This linkage is introduced by a series of progressively intimate associations between homologs . Homologs identify and pair with their unique partner . The assembly of a proteinaceous structure , the synaptonemal complex ( SC ) , stabilizes this pairing in a process called synapsis . In the context of synapsis , some programmed double strand breaks ( DSBs ) are selectively repaired to form crossovers , which give rise to chiasmata . However , how these events are coordinated to guarantee that each homolog pair gets at least one chiasma ( the obligate crossover ) is poorly understood . In all meiotic organisms studied , DSBs outnumber crossovers ( CO ) and can be repaired through multiple mechanisms [1] , [2] . Therefore , crossover formation is regulated at multiple points during meiotic DNA repair to assure the obligate crossover . Once a DSB forms , it can either be repaired using the homolog or the sister chromatid as a template . Meiotic DNA repair is biased towards interhomolog recombination to promote the formation of chiasmata [3] . Repair intermediates are then routed through crossover ( CO ) or non-crossover ( NCO ) pathways . This decision is thought to be made early , soon after DSB formation [4] , and is inhibited by the presence of nearby crossovers ( or CO-eligible intermediates ) , a phenomenon known as crossover interference [5] . Next , a subset of CO-eligible intermediates become crossovers , a process called designation [6] . In addition to these levels of control , feedback or surveillance mechanisms appear to respond to the absence of CO-eligible intermediates [7]–[9] . In C . elegans , continued phosphorylation of the nuclear envelope protein SUN-1 is associated with defects in synapsis and recombination , suggesting that this post-translational modification is an indicator of the activation of such a surveillance or feedback mechanism [8] , [10] . The role of the conserved meiotic AAA+-ATPase Pch2/TRIP13 has been controversial . In both budding yeast and mice , the PCH2/Trip13 gene has been implicated at various points in this recombination pathway . Budding yeast pch2 mutants exhibit elevated rates of DNA repair from sister chromatids [11] , [12] , misregulation of CO interference in some genetic intervals and an inability to buffer a reduction in DSBs [13] , [14] . This is in addition to other reported defects in meiotic chromosome structure [13] , [15] and DSB formation [16] , [17] . In mice , the requirement for Trip13 in meiotic chromosome metabolism is conserved [18] and cytological analysis reveals defects in DNA repair and CO formation , distribution and interference in mice deficient in Trip13 function [19] , [20] . Work in Drosophila has identified PCH2 as a component of a checkpoint that responds to defects in recombination and meiotic chromosome structure [21] , [22] . However , in C . elegans , PCH-2 was identified as a component of a checkpoint that monitors synapsis independent of defects in recombination [23] . PCH-2 is conserved in organisms that undergo synapsis during meiosis [24] , raising the possibility that the gene product is involved in this process . AAA-ATPases typically couple ATP hydrolysis to the disassembly of macromolecular complexes [25] . However , the identity of a PCH-2 substrate has remained elusive . Recently , biochemical analysis has revealed that Pch2 purified from budding yeast is capable of removing the meiotic chromosomal protein Hop1 from DNA in an ATP-dependent reaction [26] . Hop1 is a member of the HORMA domain containing family of proteins [27] and is required for pairing , synapsis and inter-homolog recombination [28]–[32] . In vivo , Pch2 may remodel Hop1 on meiotic chromosomes to promote these events [26] . We show that PCH-2 is required to restrain pairing , synapsis and recombination during meiotic prophase . Synapsis and recombination occur more rapidly in pch-2 mutants than in wildtype worms , producing subtle meiotic defects , and loss of pch-2 suppresses defects in pairing , synapsis and recombination in some mutant backgrounds . PCH-2 is expressed in germline nuclei prior to meiotic entry . Once meiosis initiates and chromosomes are synapsed , PCH-2 localizes to the SC when homolog-dependent repair mechanisms are active in the worm germline and loss of PCH-2 results in premature loss of homolog-dependent repair mechanisms in mid-pachytene . We conclude that PCH-2 inhibits meiotic prophase events by introducing a kinetic barrier to pairing , synapsis and recombination , potentially by destabilizing their intermediates . In support of this model , defects in synapsis and recombination are less frequent when pch-2 mutants are incubated at lower temperature and more frequent in both wildtype and pch-2 mutant worms grown at higher temperature . Since mutation of pch-2 results in more severe meiotic defects when combined with mutations that abrogate germline apoptosis or mildly affect meiotic DNA repair , we hypothesize that PCH-2 restrains meiotic prophase events to coordinate them , promoting quality control and preventing meiotic defects . We discuss how these findings contribute to our understanding of PCH-2 as a checkpoint protein .
To determine what role PCH-2 plays in regulating meiotic prophase events in C . elegans , we analyzed pairing , synapsis and recombination in the hermaphrodite germline . Meiotic nuclei are arrayed in a spatiotemporal gradient in the germline , allowing for the analysis of the progression of meiotic events as a function of position in the germline ( see cartoon in Figure 1A ) . We divided germlines into six zones of equal size and evaluated the steady-state levels of pairing of the left end of the X chromosome ( as visualized by the binding of the protein HIM-8 ) [33] and the 5S rDNA ( as visualized by FISH ) in each of these zones and observed no difference in the pairing of these two loci in wildtype and pch-2 single mutants ( Figure S1 ) . We analyzed SC assembly in wildtype and pch-2 mutants . The SC is assembled in two steps: axial element proteins , such as HTP-3 [34] , [35] , load prior to synapsis , and central element components , such as SYP-1 [36] , load concomitant with synapsis . Nuclei that have assembled full SC between all chromosome pairs exhibit complete colocalization of HTP-3 and SYP-1 while nuclei with incomplete synapsis have stretches of HTP-3 without SYP-1 ( nucleus indicated by arrows in Figure 1B ) . We calculated the percentage of nuclei with complete synapsis as a function of meiotic progression . Meiotic nuclei in wildtype hermaphrodites grown at 20°C initiated synapsis in the transition zone , corresponding to zone 2 , and 99% of meiotic nuclei had completed SC assembly by zone 4 ( Figure 1Cii ) . pch-2 mutants grown at 20°C also initiated SC assembly in zone 2 but twice as many nuclei completed SC assembly in this early stage of meiotic prophase ( Figure 1Cii ) . Furthermore , nuclei with complete synapsis peaked at 95% in pch-2 mutants ( zones 4 and 5 ) , indicating some slight defects in SC assembly in this mutant background ( Figure 1Cii ) . In zone 6 , SC disassembly was also disrupted in pch-2 mutants grown at 20°C when compared to wildtype worms ( Figure 1Cii ) . Therefore , pch-2 is required to restrain synapsis early in meiotic prophase and promote SC disassembly in late meiotic prophase when these events occur at 20°C . In budding yeast , temperature can modify pch2 mutant phenotypes [13] . We investigated what effect temperature might have on SC assembly and disassembly in pch-2 mutants by monitoring synapsis at 15°C and 25°C . At 15°C , pch-2 mutants accelerated synapsis , albeit less dramatically than at 20°C , and exhibited no defects in assembling or disassembling the SC ( Figure 1Ci ) . By contrast , nuclei in pch-2 mutants grown at 25°C completed SC assembly at frequencies similar to wildtype meiotic nuclei ( Figure 1Ciii ) . The similar progression of SC assembly in wildtype and pch-2 mutants was accompanied by an increase in nuclei with unsynapsed chromosomes in zones 3 and 4 of the germline in both genotypes ( Figure 1Ciii ) . By zone 4 , when the percentage of nuclei with complete synapsis peaked in both wildtype and pch-2 mutants , 16% and 11% of meiotic nuclei had unsynapsed chromosomes in wildtype and pch-2 mutant worms , respectively . pch-2 mutants also disassembled their SCs less efficiently at higher temperature . These data indicate that we can uncouple the acceleration of SC assembly from the defect in SC disassembly by incubating pch-2 mutants at lower or higher temperature . We dissected how recombination was affected by loss of pch-2 at different temperatures . First , we assessed the progress of meiotic DNA repair by localizing the recombination protein , RAD-51 , in the germlines of wildtype and pch-2 mutants grown at 15°C , 20°C and 25°C ( Figures 2A and B ) . RAD-51 is required for DNA repair and its presence on chromosomes indicates the introduction of DSBs while its disappearance indicates progression of DSBs through a repair pathway [37] . We determined the average number of RAD-51 foci per nucleus as meiotic nuclei progressed through the germline . At both 15°C and 25°C , average number of RAD-51 foci were similar between wildtype and pch-2 mutants ( Figure 2Bi and iii ) . However , pch-2 mutants grown at 20°C had significantly fewer average number of RAD-51 foci in meiotic nuclei zones 4 and 5 than wildtype worms grown at the same temperature ( Figure 2Bii ) . The reduction in average number of RAD-51 foci on chromosomes in pch-2 mutants at 20°C could either be the product of a reduction in DSBs or more rapid repair of DSBs . To distinguish between these possibilities , we assayed RAD-51 foci in rad-54 and rad-54;pch-2 double mutants ( Figure S2 ) . Mutation of rad-54 prevents the removal of RAD-51 from repair intermediates and stalls the progression of meiotic recombination [38] . rad-54 single mutants and rad-54;pch-2 double mutants exhibited a similar profile of RAD-51 loading , both in terms of kinetics and number of foci ( Figure S2 ) , indicating that the reduction in average number of RAD-51 foci we quantified in pch-2 single mutants at 20°C was the product of more rapid repair of DSBs and not a decrease in the introduction of DSBs . We evaluated how these differences in meiotic DNA repair affected crossover formation . We used GFP::COSA-1 as a cytological reporter to monitor putative crossover formation [6] in wildtype and pch-2 mutants grown at different temperatures ( Figure 2C ) . Most nuclei in wildtype worms exhibit six GFP::COSA-1 foci ( Figure 2D ) , corresponding to the six pairs of homologous chromosomes that each undergo one crossover [6] . This was particularly striking at 15°C: no nuclei had fewer than six GFP::COSA-1 foci . At 20°C and 25°C , a low percentage of meiotic nuclei in wildtype worms had fewer than six GFP::COSA-1 foci ( 2 . 3% and 1 . 5% respectively ) ( Figures 2C and E ) . pch-2 mutants exhibited a greater range of nuclei with fewer than six GFP::COSA-1 foci . At 15°C , pch-2 mutants had none , similar to wildtype worms ( Figure 2C ) . However , at 20°C , pch-2 mutants had a significantly higher fraction of meiotic nuclei with fewer than six GFP::COSA-1 foci than wildtype worms , indicating that crossover assurance mechanisms were compromised in this mutant background . This difference between wildtype and pch-2 mutants became less severe at 25°C . To investigate the effects on chiasmata formation , we identified the number of achiasmate chromosomes in wildtype and pch-2 mutants at each temperature ( arrows in Figure 2F ) . Consistent with our analysis of GFP::COSA-1 foci , there were no achiasmate chromosomes in wildtype and pch-2 mutants grown at 15°C ( Figure 2G ) . We also did not observe achiasmate chromosomes in wildtype worms grown at 20°C ( Figure 2G ) . pch-2 mutants grown at 20°C and 25°C had a small percentage of achiasmate chromosomes ( 2 . 4% and 1 . 9% respectively ) , as did wildtype worms grown at 25°C ( 1% ) ( Figure 2G ) . Thus , our analysis of meiotic DNA repair and crossover formation indicated that performing these tasks at lower temperature suppressed defects observed in pch-2 mutants grown at 20°C and 25°C . Wildtype worms exhibited defects in these events when incubated at 25°C ( Figure 2G ) . Since both synapsis and meiotic DNA repair were accelerated in pch-2 mutants grown at 20°C , we tested whether meiotic nuclei in pch-2 mutants entered meiosis earlier or progressed through the germline more rapidly than in wildtype worms at 20°C . Phosphorylation of the nuclear envelope protein SUN-1 occurs at meiotic entry [10] , [39] and persists until synapsis is complete and meiotic recombination is proceeding normally [10] . We localized SUN-1 phosphorylated on serine 8 ( SUN-1 pSer8 ) in wildtype and pch-2 mutants grown at 20°C and did not detect any difference in its appearance or disappearance between the two genotypes ( Figure 3 ) . To determine if nuclei were progressing through the germline more rapidly , we EdU labeled [40] meiotic nuclei in both wildtype and pch-2 mutants ( Figure S3A ) and visualized their progression through the various stages of meiotic prophase . Both wildtype and pch-2 mutants grown at 20°C exhibited a similar progression of EdU labeled nuclei throughout the timecourse ( Figures S3B and C ) . Both of these assays lead us to conclude that the effects on synapsis and recombination that we observed in pch-2 mutants were not merely the result of more rapid meiotic entry or progression of nuclei through the germline but that PCH-2 is specifically required to restrain these events . In summary , we report accelerated rates of synapsis and meiotic DNA repair , accompanied by subtle defects in synapsis and recombination , in pch-2 mutants grown at 20°C ( Figures 1Cii and 2Bii ) . Synapsis and meiotic DNA repair were less affected in pch-2 mutants cultured at 15°C ( Figures 1Ci and 2Bi ) and these mutant worms did not have any recombination or synapsis defects ( Figures 1Ci , 2C and 2G ) . Any difference in the rate of SC assembly and meiotic DNA repair between wildtype and pch-2 mutants was lost at 25°C ( Figures 1Ciii and 2Biii ) and this correlated with an increase in the frequency of synapsis and recombination defects in wildtype worms ( Figures 1Ciii , 2C and 2G ) . Moreover , defects in SC disassembly in pch-2 mutants correlated with defects in meiotic recombination ( Figures 1Cii , 1Ciii , 2C and 2G ) . Our previous experiments indicated that mutating pch-2 had subtle effects on synapsis and recombination . To more completely dissect how PCH-2 might be affecting homolog interactions , we investigated the effect of mutating pch-2 in sensitized mutant backgrounds . First we analyzed pairing in syp-1 and syp-1;pch-2 double mutants . syp-1 mutants fail to assemble SC between paired homologs , allowing the visualization of pairing intermediates that typically precede and promote synapsis [35] , [36] . In C . elegans , cis-acting sites called Pairing Centers ( PCs ) are required for efficient pairing and synapsis [35] . In the absence of synapsis , homologous chromosomes exhibit stable pairing of PC ends but not non-PC ends of chromosomes [36] . This interaction depends on both homologs having functional PCs [35] . We monitored synapsis-independent , PC-dependent pairing of X chromosomes as a function of meiotic progression by performing immunofluorescence against HIM-8 [41] ( Figure 4A ) . In both syp-1 single mutants and pch-2;syp-1 double mutants , PCs of X chromosomes initiated pairing in zone 2 and maintained this pairing until late meiotic prophase ( zone 6 ) ( Figure 4B ) . Interestingly , the PC end of X chromosomes was more frequently paired in pch-2;syp-1 double mutants in zone 2 than syp-1 single mutants , at levels similar to wildtype in the same region of the germline ( Figure 4B ) . Therefore , loss of pch-2 suppressed the pairing defect of syp-1 mutants early in prophase . To determine if this effect was an indirect consequence of a reduction in apoptosis affecting the overall population of meiotic nuclei inhabiting hermaphrodite germlines , we also monitored X chromosome PC pairing in spo-11;syp1 double mutants , which display a similar level of germline apoptosis as pch-2;syp-1 mutants due to abrogation of the DNA damage checkpoint [23] . This double mutant background did not display more stable association of X chromosome PCs ( Figure 4B ) , indicating that PCH-2 destabilizes synapsis-independent , PC-dependent pairing . Loci that are located some distance away from PCs also exhibit synapsis-independent pairing , although to a lesser extent than that observed at PCs [36] . To evaluate the effect of mutating pch-2 on pairing of an autosomal , internally located locus , we performed FISH against the 5 s rDNA locus ( Figure 4C ) . In syp-1 single mutants , there was a gradual increase in pairing of this locus until zone 3 ( Figure 4D ) . Once again , in pch-2;syp-1 double mutants , we observed higher levels of pairing at this locus in both zones 2 and 3 ( Figure 4D ) , indicating that PCH-2 also negatively regulates pairing at a non-PC locus . Given that mutation of pch-2 affected homolog interactions at both PC and non-PC loci , we determined what effect loss of pch-2 would have on synapsis that could not initiate at two stably paired PCs . meDf2 is a deficiency that removes the X chromosome PC [35] , [42] . Animals heterozygous for meDf2 ( meDf2/+ ) exhibit unsynapsed X chromosomes in 50% of meiotic nuclei [35] ( Figure 5B ) , illustrating that homolog synapsis can occur , albeit inefficiently , when only one chromosome has a functional PC . We tested whether mutation of pch-2 in meDf2 heterozygotes had any effect on the synapsis defect in these hermaphrodites . When we monitored synapsis as a function of meiotic progression in meDf2/+ animals , there was a gradual increase in meiotic nuclei with complete colocalization of HTP-3 and SYP-1 , which mirrored the gradual increase in pairing that has been reported for the X chromosome in meDf2 heterozygotes [35] . By zone 5 , 48% of meiotic nuclei had completely assembled SC between all homolog pairs while the remainder exhibit stretches of HTP-3 without SYP-1 ( arrows in Figure 5A ) , which are the unsynapsed X chromosomes ( Figure 5C ) . pch-2;meDf2/+ double mutants also exhibited a gradual increase in the percentage of meiotic nuclei with complete synapsis ( Figures 5A and B ) . However , in this double mutant background , most meiotic nuclei ( 87% ) achieved complete synapsis in zone 5 ( Figure 5B ) , indicating that mutation of pch-2 suppressed the synapsis defect of meDf2 heterozygotes . These data explain the reduction in apoptosis we previously reported in meDf2/+;pch-2 double mutants when compared to meDf2/+ single mutants [23] . We verified that synapsis was occurring between homologous X chromosomes by performing FISH against a non-PC locus in pch-2;meDf2/+ double mutants ( Figures 5C and D ) . We tested whether the synapsis in pch-2;meDf2/+ double mutants was functional for crossover formation by analyzing the number of nuclei with achiasmate chromosomes in meDf2/+ single mutants and pch-2;meDf2/+ double mutants ( Figure 5E ) . Introduction of the pch-2 mutant allele suppressed the appearance of achiasmate chromosomes in meDf2 heterozygotes ( Figure 5E ) , indicating that synapsis in pch-2;meDf2/+ double mutants can support crossover formation . We next evaluated the percentage of male-self progeny produced by these double mutants . Since males ( with a single X chromosome ) typically arise from spontaneous non-disjunction of X chromosomes at a low frequency in wildtype hermaphrodites ( with two X chromosomes ) , mutations that increase the frequency of X chromosome non-disjunction will exhibit an increase in male self-progeny . For example , meDf2/+ hermaphrodites produced 11 . 6% male progeny ( Table 1 ) . Consistent with our analysis of synapsis and recombination in pch-2;meDf2/+ double mutants , these hermaphrodites produced 0 . 9% male self-progeny ( Table 1 ) , indicating that X chromosomes segregated correctly during meiosis . Altogether , these data provide additional evidence that PCH-2 normally acts to restrain synapsis , even in situations in which synapsis cannot initiate from two stably paired PCs [35] . Thus far , the acceleration of pairing , synapsis and DNA repair in pch-2 mutants suggested a role for the gene product in coordinating these events , potentially to promote quality control . If this model is correct , we reasoned that loss of both pch-2 gene function and other known mechanisms that contribute to quality control during meiotic prophase might increase the frequency of meiotic defects . For example , germline apoptosis removes defective nuclei in late meiotic prophase to prevent the production of aneuploid gametes [23] . To test our hypothesis , we introduced a mutation in ced-4 , a gene that encodes a core component of the apoptotic machinery that is required for germline apoptosis [43] , into pch-2 mutants and monitored the number of nuclei with achiasmate chromosomes . ced-4 single mutants had achiasmate chromosomes in 1% of their late prophase nuclei ( Figure 6 ) . This frequency of nuclei with achiasmate chromosomes corresponds with the frequency of wildtype meiotic nuclei with less than six GFP::COSA-1 foci ( 2% ) , consistent with germline apoptosis culling defective nuclei even during wildtype meiosis . 8% of nuclei in pch-2;ced-4 double mutants had achiasmate chromosomes ( Figure 6 ) , indicating that significantly more meiotic nuclei in pch-2 mutants have meiotic defects than was represented by the frequency of nuclei with achiasmate chromosomes in the single mutant . Moreover , the increase in meiotic nuclei with achiasmate chromosomes pch-2;ced-4 double mutants was accompanied by an increase in the production of inviable progeny , indicating the missegregation of chromosomes during meiosis ( Table 1 ) . We wanted to further test our hypothesis that PCH-2 was involved in quality control during meiotic prophase by introducing the pch-2 mutation into another mutant background to determine if the double mutants would exhibit a more severe meiotic phenotype . However , since pch-2 mutants suppress some meiotic phenotypes ( Figures 4 and 5 ) , we reasoned that we should use a mutation in which the meiotic phenotype was relatively mild , indicating no major defects in meiosis . In this scenario , loss of pch-2 could reveal a more severe meiotic phenotype . HORMA domain containing proteins promote many events during meiotic prophase , including pairing , synapsis and recombination between homologous chromosomes [34] , [44]–[48] . There are four genes in the C . elegans genome that encode meiotic HORMA domain containing proteins ( him-3 , htp-1 , htp-2 , and htp-3 ) and recently a hypomorphic allele of htp-3 , vc75 , has been identified . Worms with this mutation appear to undergo meiosis normally , producing no achiasmate chromosomes ( Figure 6 ) , but exhibit minor defects in DNA repair and SC disassembly [49] . We generated pch-2;htp-3 ( vc75 ) double mutants and monitored the viability of its progeny and chiasmata formation in these double mutants . pch-2;htp-3 ( vc75 ) double mutants had a synthetic meiotic phenotype: Although each individual mutant produced no inviable progeny , the double mutant generated 20% inviable progeny ( Table 1 ) . This reduction in viability can be explained by the significant increase in achiasmate chromosomes ( Figure 6 ) , indicating that pch-2;htp-3 ( vc75 ) mutants had a more severe defect in recombination than either single mutant . We ruled out that this synthetic meiotic phenotype could be explained by the possibility that pch-2 mutants activate an HTP-3 dependent meiotic DNA damage response by analyzing SC disassembly in double mutants . When mutations that activate an HTP-3 dependent meiotic DNA damage response are combined with the htp-3 ( vc75 ) mutant allele , these double mutants disassemble their SCs more rapidly than the htp-3 ( vc75 ) single mutant [49] . This did not occur in pch-2;htp-3 ( vc75 ) double mutants ( data not shown ) . Thus , introduction of the pch-2 mutation exacerbates the meiotic phenotype of a weak mutant allele of htp-3 , consistent with a role for PCH-2 in quality control during meiotic prophase . To gain a deeper understanding of how PCH-2 might be regulating pairing , synapsis and recombination , we generated a polyclonal antibody against the PCH-2 protein and localized it in wildtype worms . In wildtype meiotic nuclei , PCH-2 was expressed in germline nuclei before they adopted the polarized morphology that is indicative of meiotic entry , termed the transition zone ( Figures 7A and B ) . In these nuclei , PCH-2 formed foci that did not colocalize with axial elements , such as HTP-3 [34] , [35] ( data not shown ) , or with HIM-8 [33] ( data not shown ) . After meiotic entry , PCH-2 localized to meiotic chromosomes in a manner reminiscent of proteins that compose the SC ( Figure 7A and B ) . PCH-2 was loaded into chromosomes in the transition zone ( Figures 7A and B ) and present in the portion of the worm germline that corresponds to pachytene ( Figure 7A ) . However , unlike SC proteins , PCH-2 was absent from meiotic chromosomes in the most distal region of the germline , corresponding to late pachytene ( Figure 6A ) . To confirm this localization pattern , we co-stained wildtype meiotic nuclei with antibodies against PCH-2 and the SC component SYP-1 [36] . In mid-pachytene , SYP-1 and PCH-2 colocalized ( Figure 7C ) , consistent with PCH-2 localizing to the SC . In late pachytene , SYP-1 was present on meiotic chromosomes but PCH-2 was not ( Figure 7D ) , indicating it was removed from the SC before the SC disassembled . We tested the genetic requirements for PCH-2 loading on meiotic chromosomes . PCH-2 was not present in germline nuclei when pch-2 was inactivated by RNA interference in wildtype worms ( data not shown ) . Synapsis is required for PCH-2 localization to the SC , since PCH-2 localization was lost in syp-1 mutants [36] ( Figure 7E ) . However , homologous synapsis was not , as illustrated by PCH-2's localization to the SC in htp-1 ( Figure 7F ) and sun-1 mutants ( data not shown ) , in which chromosomes synapse non-homologously [46] , [50] , and in ieDf2 mutants ( data not shown ) , in which chromosomes self-synapse [51] . PCH-2 was present on the SC when meiotic recombination is disrupted , such as in msh-5 [52] , zhp-3 [53] , spo-11 [54] ( Figure S5 ) , or rtel-1 mutants [55] , [56] ( data not shown ) . We also determined that htp-3 ( vc75 ) mutants and wildtype animals grown at 25°C exhibited normal localization of PCH-2 ( data not shown ) . The region of the germline where PCH-2 was absent from meiotic chromosomes is associated with a shift in double strand break ( DSB ) repair from use of the homolog as a repair template , presumably to use of the sister chromatid [57] , [58] . This shift in DNA repair mechanisms is contemporaneous with the loading of the crossover-promoting factor , COSA-1 [6] . To verify that PCH-2's absence from meiotic chromosomes coincided with these events , we performed immunofluorescence on wildtype hermaphrodites harboring a GFP tagged COSA-1 transgene [6] with antibodies against PCH-2 . PCH-2 was indeed absent from nuclei that had robust GFP::COSA-1 foci ( Figure S4 ) . The shift in DNA repair mechanisms is dependent on the activity of the MAP kinase pathway [57] , the crossover-promoting factor , msh-5 , and the anti-recombinase , rtel-1 [58] . PCH-2's removal from meiotic chromosomes in late pachytene did not occur in mpk-1 ( Figure 7G ) and msh-5 mutant backgrounds ( Figures 7I and S5 ) but did in rtel-1 mutants ( Figure 7J ) . This relocalization also failed to occur when we alleviate the meiotic progression defect in mpk-1 mutants by RNAi of a downstream effector , dpl-1 [59] ( Figure 7H ) , indicating that an inability to remove PCH-2 from meiotic chromosomes is not merely an indirect consequence of meiotic prophase arrest in mpk-1 mutants . PCH-2 was also present on the SC in late pachytene nuclei in spo-11 and zhp-3 mutants , suggesting that PCH-2's perdurance is a general response to defects in recombination , similar to the persistence of phosphorylated SUN-1 [10] ( Figure S5 ) . The timing of PCH-2 localization and removal from the SC ( Figures 7C and D ) , along with the effects that mutation of pch-2 had on meiotic recombination ( Figure 2 ) , led us to speculate that PCH-2 might specifically be required to promote homolog dependent DNA repair . To test this , we first monitored the dynamics of RAD-51 loading and removal in syp-1 and pch-2;syp-1 mutants ( Figure 8A ) . The complete absence of SC between all homologous chromosomes in syp-1 mutants [36] results in an inability to repair programmed DSBs because of the absence of closely aligned homologous chromosomes that can be used as repair templates [37] . As a result , RAD-51 foci persist until late pachytene ( zone 6 in Figure 8A ) , when a synapsis-independent repair program is activated . When we compared syp-1 and pch-2;syp-1 mutants , we observed that mutation of pch-2 reduced the average number of RAD-51 foci per nucleus in syp-1 mutants in zone 5 ( Figure 8A ) . Since our previous experiments ruled out the possibility that mutation of pch-2 reduced the number of DSBs ( Figure S1 ) , these data indicated that some DSBs were being repaired in pch-2;syp-1 double mutants by synapsis-independent mechanisms , presumably by using the sister chromatid as a repair template . Despite this premature activation of synapsis-independent DSB repair mechanisms , pch-2;syp-1 double mutants exhibited an extended transition zone [36] and prolonged zone of meiotic nuclei with phosphorylated SUN-1 [10] , indicating that meiotic progression is unaffected by mutation of pch-2 in syp-1 mutants ( data not shown ) . Moreover , our data also raise the possibility that the partial suppression of recombination defects in pch-2;syp-1 double mutants is temporally constrained to mid-pachytene ( zone 5 ) since the average number of RAD-51 foci was not significantly different between syp-1 and pch-2;syp-1 mutants in earlier zones of the germline ( Figure 8A ) . We directly tested the role of PCH-2 in promoting homolog-dependent meiotic DNA repair . We used a meiotic recombination assay recently developed to monitor when meiotic chromosomes can access their homolog for DNA repair [58] . This assay takes advantage of an insertion of the Mos1 transposon into the unc-5 locus on one homologous chromosome , which disrupts the gene . DSBs are generated in young adult hermaphrodites carrying this locus when heat shock induced expression of the Mos transposase excises the Mos1 transposon . Since the other homologous chromosome contains a mutation that inactivates the unc-5 gene elsewhere in the gene , the outcome of DNA repair produces a wildtype unc-5 ( + ) allele , allowing these recombinants to be phenotypically identified among the progeny laid by heat-shocked worms . The mode of DNA repair can be determined by monitoring the progeny of unc-5 ( + ) recombinants , since the presence of a linked marker ( dpy-13 ) can distinguish non-crossover from crossover recombinants ( Figure 8B ) . Given the spatiotemporal organization of meiotic nuclei in the hermaphrodite germline , recombinants laid at different time points reflect the competence of meiotic chromosomes to use their homolog for DNA repair at different stages of meiotic prophase . When we performed this assay with wildtype and pch-2 mutant hermaphrodites , we did not recover any recombinants in the first time point ( 10–22 hours post-heat shock ) in both genetic backgrounds ( Figure 8C ) , which corresponds to late pachytene in the hermaphrodite germline when homolog-dependent repair is not active [57] , [58] . Given the subtle defects in pairing , synapsis and recombination we observed in pch-2 mutants ( Figures 1 , 2 and 3 ) , we divided the next twelve-hour time point ( 22–34 hours post heat shock ) into two six hour timepoints ( 22–28 hours post heat shock and 28–34 hours post heat shock ) . Wildtype animals produced recombinant progeny in the 22–28 hour window after heat shock at a frequency of 4% , reflecting the increased competence of nuclei in mid-pachytene for DNA repair from a homologously synapsed partner . In pch-2 mutants , statistically significant fewer recombinants were generated during this period and of the nine recombinants identified , none were crossovers . In wildtype worms , 21% of recombinants recovered at this timepoint were crossovers . The difference in the frequency of recombinants between wildtype and pch-2 mutants was lost in later timepoints ( Figure 8C ) . By the next six-hour interval ( 28–34 hours post heat-shock ) , wildtype hermaphrodites generated recombinants at a frequency of 5% and pch-2 mutants at a frequency of 4% . Wildtype and pch-2 mutant hermaphrodites generated equivalent frequencies of recombinants until the end of the timecourse . Thus , PCH-2 is required to maintain access to homologous chromosomes for meiotic DNA repair during pachytene . To determine what effect premature loss of homolog access had on crossover formation in pch-2 mutants , we monitored recombination genetically in wildtype and pch-2 mutant animals using five single nucleotide polymorphisms ( SNPs ) that spanned 95% of Chromosome III or the X chromosome [60] ( Figure 8D ) . pch-2 mutants had slightly lower rates of recombination in all intervals across both chromosomes , resulting in a reduction of the genetic length of both Chromosome III ( from 55 . 4 to 43 . 1 cM ) and the X chromosome ( from 58 . 5 to 45 . 8 cM ) . However , the decrease in genetic distance between SNPs was less severe at the PC end of both chromosomes . In addition , mutation of pch-2 appeared to more strongly affect the genetic distance between SNPs in the center of Chromosome III . We observed several double crossovers in wildtype animals on both Chromosome III and the X chromosome , 8 and 10 respectively . In contrast to wildtype , pch-2 mutants had significantly fewer double crossovers , only 2 on the X chromosome and none on Chromosome III . Therefore , the acceleration of homolog-dependent DNA repair in pch-2 single mutants at 20°C , as assayed both by the appearance and disappearance of RAD-51 foci ( Figure 2Bii ) and the ability of chromosomes to access their homolog for DNA repair ( Figure 8C ) , reduced the genetic length of chromosomes , albeit not uniformly among genetic intervals , and the frequency of double crossovers . The localization pattern of PCH-2 in msh-5 mutants suggested that the persistence of PCH-2 on meiotic chromosomes into late pachytene might be a cytological read-out for prolonged homolog access and that PCH-2 might be required for this extension . However , we could not directly test this possibility in pch-2;msh-5 double mutants since the assay to monitor homolog access relies on large numbers of viable progeny to assess significance [58] . Instead , we chose to address this possibility in meDf2 homozygotes . In this mutant , asynapsis of X chromosomes produces changes in recombination frequencies in some genetic intervals and a loss of crossover control so that autosomes sometimes experience double crossovers [61] . This redistribution of recombination events in meDf2 homozygotes could , in part , be the consequence of PCH-2 remaining on meiotic chromosomes in late pachytene and promoting inter-homolog recombination in response to defects in synapsis . First we determined how PCH-2 localization was affected by asynapsis of X chromosomes and whether any change in its localization correlated with characterized responses to asynapsis . We stained meDf2 hermaphrodite germlines with antibodies against PCH-2 and phosphorylated SUN-1 [10] . The staining pattern for both was extended ( Figure 9A ) , in contrast to their staining patterns in wildtype germlines ( Figures 3 and 7A ) . This extension was accompanied by a delay in the appearance of GFP::COSA-1 foci ( data not shown ) . Next , we analyzed the number of GFP::COSA-1 foci per nucleus [6] ( Figure 9B ) . In meDf2 mutants , the majority of germline nuclei ( 58% ) exhibited five GFP::COSA-1 foci . However , a substantial fraction of meiotic nuclei ( 34% ) had six GFP::COSA-1 foci . Even taking into account that 10% of these nuclei had undergone successful synapsis of X chromosomes [35] , this suggested that about one-quarter of meiotic nuclei had an autosome with an additional putative crossover event . When we analyze GFP::COSA-1 foci in pch-2;meDf2 double mutants , we observed a statistically significant reduction in the number of germline nuclei that have six foci ( 22% ) and a corresponding increase in the number of nuclei with five foci ( 74% ) . Mutation of pch-2 had no effect on the percent of nuclei with complete synapsis in homozygous meDf2 mutants ( data not shown ) . These results suggest that cytologically , PCH-2 is required for the increase in putative double crossovers in meDf2 mutants . However , nuclei with six GFP::COSA-1 foci were still present in pch-2;meDf2 double mutants , indicating that additional mechanisms promote homolog-dependent meiotic DNA repair in meDf2 mutants . This possibility was supported by the similar average numbers of RAD-51 foci in both meDf2 and meDf2;pch-2 mutants ( Figure 9C ) . We also assessed double crossover formation genetically in both meDf2 and pch-2;meDf2 double mutants on Chromosome III . Mutation of pch-2 did not ameliorate the minor shifts in recombination pattern that we observed in meDf2 . Indeed , when represented as fractions of wildtype recombination rates , the recombination landscape was more severely affected in pch-2;meDf2 double mutants than meDf2 single mutants . There was no difference in the frequency of double crossovers between meDf2 and meDf2;pch-2 ( Figure 9D ) , likely because of the removal of meiotic nuclei by germline apoptosis ( see Discussion ) .
PCH-2 is required to maintain interhomolog meiotic recombination ( Figure 8 ) . A similar role for Pch2 has been demonstrated in budding yeast [11] , [12] . However , our experiments reveal that the requirement for PCH-2 in promoting interhomolog meiotic recombination is temporally constrained to mid-pachytene ( Figures 8A and C ) . We cannot interpret any other differences between wildtype and pch-2 mutants at later timepoints ( namely the 58–70 hours phs and 70+ hours phs timepoints ) since Mos transposase may persist after induction and continue to introduce DSBs throughout the germline . Mos-induced DSBs can be repaired by inter-homolog repair in pch-2 mutants prior to this window in mid-pachytene , as indicated by the frequency of recombinants laid soon after the 22–28 hour timepoint ( Figure 8C ) . Thus , PCH-2 appears to be redundant with additional mechanisms that promote inter-homolog meiotic recombination earlier in prophase . In addition , pch-2 mutants do not load GFP::COSA-1 prematurely ( data not shown ) , indicating that switching from meiotic inter-homolog DNA repair to a homolog-independent DNA repair mode can be uncoupled from crossover designation . Alternatively , designation may occur earlier , concomitant with loss of homolog access , and visible COSA-1 recruitment stabilizes the designated crossover . Our model of how PCH-2 regulates meiotic recombination is illustrated in Figure 10 . Nuclei begin introducing DSBs coincident with the localization of DSB-1 and DSB-2 , proteins required for DSB formation and thought to identify a DSB competent period during prophase [8] , [9] . A subset of these DSBs become licensed as CO-eligible recombination intermediates [6] . Nuclei in which all chromosomes have a CO-eligible intermediate progress through mid-pachytene and a subset of these CO-eligible intermediates are designated as crossovers in late pachytene [6] . In a small population of nuclei that contain a chromosome without a CO-eligible intermediate , 5–10% based on our studies of GFP::COSA-1 in pch-2 mutants ( Figure 2C ) , DSB competence is thought to be prolonged to increase the likelihood that a CO-eligible intermediate will be formed . This extension is accompanied by PCH-2-dependent promotion of interhomolog recombination , providing a potential explanation for the presence of double crossovers amongst the progeny of wildtype animals that are lost in the progeny of pch-2 mutants ( Figure 8D ) . Why is there a mechanism unique to mid-pachytene to maintain homolog access ? Our model assumes the requirement for PCH-2 in mid-pachytene is linked to the maintenance of DSB competence when a chromosome does not have a CO-eligible intermediate . Some possibilities that could explain our results are: 1 ) The redundant homolog-dependent repair mechanism ( s ) active earlier in prophase is down-regulated in mid-pachytene , even if DSB competence is maintained . We do not favor this possibility ( see below ) . 2 ) There is something unique about DSBs or the environment in which DSBs are introduced during mid-pachytene that requires additional reinforcement to promote the formation of crossovers . These models are not mutually exclusive . We favor another interpretation . Since nuclei in mid-pachytene continue to localize PCH-2 to meiotic chromosomes but do not exhibit other molecular markers associated with DSB competence , such as SUN-1Ser8P [7]–[9] ( Figure S6 ) , we suggest that the requirement for PCH-2 in mid-pachytene is independent of DSB competence . This possibility takes into account events that are thought to occur during mid-pachytene . CO-eligible intermediates have been licensed but they have not yet been designated . Villeneuve and colleagues have speculated that a reinforcement step precedes designation , whereby a single CO-eligible intermediate per chromosome is reinforced by CO-promoting mechanisms and ultimately stabilized by designation [6] . Since our analysis of pairing , synapsis and meiotic DNA repair indicates that PCH-2 typically restrains meiotic events , another way to interpret the loss of homolog access in pch-2 mutants is that it reveals the prematurity of another event , namely reinforcement of crossovers . Since licensed CO-eligible intermediates that are not reinforced are presumably repaired by NCO mechanisms [6] , homolog access would be maintained during this process . In the absence of PCH-2 and the more rapid reinforcement of CO-eligible intermediates , homolog access during mid-pachytene would also be lost . This could explain the redistribution of crossovers in pch-2 mutants ( Figure 8D ) . We think it unlikely that PCH-2 signals the absence of CO-eligible intermediates as its loss does not affect the staining pattern of SUN-1Ser8P [10] and DSB-2 [7]–[9] ( data not shown ) in genetic backgrounds that have defects in recombination and/or synapsis ( i . e . meDf2 and syp-1 ) . Indeed , crossover distribution in pch-2;meDf2 double mutants appears to be the additive effect of loss of pch-2 and the delay in meiotic progression introduced by asynapsis ( Figure 9D ) , suggesting that these events are independent of each other . In contrast to wildtype hermaphrodites , PCH-2 localizes to the SC of meiotic chromosomes in late pachytene when synapsis and/or recombination are defective ( Figures 7I , S5 and 9A ) . This delay in PCH-2 removal correlates with other indicators associated with a delay in meiotic prophase progression , such as phosphorylation of SUN-1 [10] ( Figures S5 and 9A ) and persistence of DSB-1 and DSB-2 [8] , [9] ( see nucleus labeled defective meiosis in Figure 10 ) . We suggest that the continued localization of PCH-2 to meiotic chromosomes in mutant backgrounds that perturb synapsis and/or recombination ( Figures S5 and 9A ) indicates the existence of a feedback mechanism that maintains inter-homolog recombination to promote crossover assurance . Such a model would explain the delay in the appearance of GFP::COSA-1 foci in meDf2 mutants ( data not shown ) and the reduction in nuclei with six GFP::COSA-1 foci in meDf2;pch-2 double mutants ( Figure 9B ) . It would also explain why the “interchromosomal effect” observed in Drosophila , the global increase in crossovers produced by chromosomal rearrangements , depends on PCH2 [22] . In contrast to our cytological analysis of GFP::COSA-1 foci , we do not see a corresponding decrease in the number of genetically observed double crossovers in meDf2;pch-2 double mutants , likely due to small number of double crossovers observed in both mutant backgrounds ( Figure 9D ) . The discrepancy between cytological and genetic evidence of double crossovers could be the result of germline apoptosis . Since the number and kinetics of RAD-51 focus formation are indistinguishable between rad-54 and rad-54;pch-2 ( Figure S2 ) , we also conclude that this feedback mechanism is distinct from one that has been proposed to prolong DSB competence in response to defects in synapsis and/or recombination [7]–[9] . PCH-2's proposed role in maintaining interhomolog repair mechanisms in response to defects in meiotic recombination potentially clarifies why PCH2 is required for crossover homeostasis [13] , [14] . A reduction in DSB formation increases the likelihood that chromosomes do not have a CO-eligible recombination intermediate and would activate described feedback mechanisms , including an attempt to increase DSBs [7]–[9] and prolong interhomolog access for repair . Since DSB formation is compromised , there is a greater reliance on maintaining homolog dependent meiotic DNA repair , a pathway ( partially ) dependent on PCH2 . This potential relationship between defective meiotic recombination and Pch2's ability to maintain inter-homolog repair into late meiotic prophase could also explain why prophase arrest rescues a reduction in DSB activity but only affects wildtype meiosis minimally [66] , especially if Pch2 activity is regulated similarly in budding yeast as in C . elegans . However , mutation of pch-2 does not eradicate nuclei with greater than five GFP::COSA-1 nuclei in the meDf2 background ( Figure 9B ) . Moreover , our analysis of RAD-51 in meD2;pch-2 ( Figure 9C ) does not indicate the activation of homolog-independent repair mechanisms , in contrast to our studies with syp-1;pch-2 mutants ( Figure 8A ) . Therefore , we also conclude that at least one other pathway in C . elegans , in addition to PCH-2 , maintains homolog access in response to defects in synapsis and recombination and that this pathway relies on synapsis for its activity . An attractive candidate for this pathway is the one that we hypothesize acts earlier in meiotic prophase and may be coupled to DSB competence [8] , [9] .
The wildtype C . elegans strain background was Bristol N2 [67] . All experiments were performed at 20° under standard conditions unless otherwise stated . Mutations and rearrangements used were as follows: LG I: mnDp66 , rtel-1 ( tm1866 ) , rad-54&snx-3 ( ok615 ) , zhp-3 ( jf61 ) , hT2[bli-4 ( e937 ) let- ? ( q782 ) qIs48] LG II: pch-2 ( tm1458 ) , meIs8 [pie-1p::GFP::cosa-1+unc-119 ( + ) ] LG III: htp-3 ( vc75 ) , dpy-5 ( e61 ) , mpk-1 ( ga111 ) , ced-4 ( n1162 ) LG IV: spo-11 ( ok79 ) , htp-1 ( gk174 ) , msh-5 ( me23 ) , dpy-13 ( e184 ) unc-5 ( ox171::Mos1 ) , unc-5 ( e791 ) , ieDf2 , nT1[unc- ? ( n754 ) let- ? ( m435 ) ] ( IV , V ) , nTI [qIs51] LG V: sun-1 ( jf18 ) , syp-1 ( me17 ) , krIs14 [hsp-16 . 48::MosTransposase; unc-122::gfp; lin-15 ( + ) ] LG X: meDf2 Some strains were provided by the CGC , which is funded by NIH Office of Research Infrastructure Programs ( P40 OD010440 ) . Polyclonal rabbit antibodies directed against the first 100 amino acids of PCH-2 were generated by Strategic Diagnostic Inc . ( SDI , Newark , DE ) . Immunostaining was performed as in [23] . Primary antibodies were as follows ( dilutions are indicated in parentheses ) : rabbit anti-SYP-1 ( 1∶500 ) [36] , guinea pig anti-SYP-1 ( 1∶500 ) [36] , guinea pig anti-HTP-3 ( 1∶500 ) [35] , chicken anti-HTP-3 ( 1∶1000 ) [35] , rabbit anti-RAD-51 ( 1∶1000 ) [37] , guinea pig anti-SUN-1S8Pi ( 1∶700 ) [68] and rabbit anti-PCH-2 ( 1∶1000 ) . Secondary antibodies were Cy3 anti-chicken , anti-guinea pig and anti-rabbit ( Jackson Immunochemicals ) , Alexa-Fluor 555 anti-rabbit , anti-guinea pig , and anti-chicken ( Invitrogen ) and Cy5 anti-guinea pig and anti-chicken ( Jackson Immunochemicals ) . Fluorescence in situ hybridization was performed as described in [33] . Quantification of synapsis , pairing and RAD-51 foci was performed with a minimum of three whole germlines per genotype . For each figure , the number of nuclei assayed for each genotype in each zone is provided in Table S1 . EdU labeling was performed as described in [58] . L4 hermaphrodites were aged for 8–10 hours and moved to plates with E . coli labeled with EdU to promote EdU labeling of C . elegans meiotic nuclei . A minimum of 15 germlines was scored for each genotype at each timepoint . All images were acquired using a DeltaVision Personal DV system ( Applied Precision ) equipped with a 100× N . A . 1 . 40 oil-immersion objective ( Olympus ) , resulting in an effective XY pixel spacing of 0 . 064 or 0 . 040 µm , and a 60× oil-immersion objective ( Olympus ) , resulting in an effective XY pixel spacing of 0 . 11 or 0 . 067 µm . Three-dimensional image stacks were collected at 0 . 2-µm Z-spacing and processed by constrained , iterative deconvolution . Image scaling and analysis were performed using functions in the softWoRx software package . Projections were calculated by a maximum intensity algorithm . Composite images were assembled and some false coloring was performed with Adobe Photoshop . The wild-type Hawaiian CB4856 strain and the Bristol N2 strain were used to assay recombination between single nucleotide polymorphisms ( SNPs ) on Chromosomes III and X . The SNPs used on Chromosome III were: pkP3081 , pkP3095 , pkP3101 , pkP3035 , and pkP3080 . The SNPs used on the X chromosome were: pkP6139 , pkP6120 , pkP6157 , pkP6161 , and pkP6170 [60] . To measure wild-type recombination , N2 males containing bcIs39 were crossed to Hawaiian CB4856 worms . Cross-progeny hermaphrodites were identified by the presence of bcIs39 and contained one N2 and one CB4856 chromosome . These were assayed for recombination by crossing with males containing mIs11 and N2 SNPs . Cross-progeny hermaphrodites from the resulting mate were isolated as L4s , and then cultured individually in 96-well plates in liquid S-media complete supplemented with HB101 , carbenicillin , and Nystatin . Four days after initial culturing , starved populations were lysed and used for PCR and restriction digest to detect CB4856 SNP alleles . For recombination in pch-2 mutants , strains homozygous for the CB4856 background of the relevant SNPs were created , then mated with pch-2; bcIs39 . Subsequent steps were performed as in the wild-type worms . The Mos1 excision-induced DSB repair assay was performed as described in [58] with the modification that the second 12 hour window ( 22–24 hours post-heat shock ) was divided into two 6 hour windows ( 22–28 post-heat shock and 28–34 post-heat shock ) . Recombinants were identified by their wildtype phenotype and their progeny analyzed to determine whether recombination was the product of a crossover or a non-crossover event . In some cases , whether the recombinants were the products of non-crossover or crossover recombination could not be determined because the recombinant hermaphrodites crawled off the agar and did not produce enough progeny .
|
The production of sperm and eggs for sexual reproduction depends on meiosis . During this specialized cell division , homologous chromosomes are linked by at least one crossover recombination event , or chiasma , to promote their proper segregation . How events in meiotic prophase are coordinated to contribute to crossover assurance is not well understood . Here , we show that C . elegans PCH-2 regulates a variety of events during meiotic prophase to promote crossover assurance . In the absence of pch-2 , pairing , synapsis and recombination are accelerated , resulting in defects in synapsis and crossover formation . We propose that PCH-2 restrains the events of meiotic prophase to coordinate them , ensure their fidelity and guarantee that each homolog pair has at least one crossover to promote proper meiotic chromosome segregation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"cell",
"biology",
"genetics",
"cell",
"biology",
"biology",
"and",
"life",
"sciences"
] |
2014
|
A Quality Control Mechanism Coordinates Meiotic Prophase Events to Promote Crossover Assurance
|
How is movement of individuals coordinated as a group ? This is a fundamental question of social behaviour , encompassing phenomena such as bird flocking , fish schooling , and the innumerable activities in human groups that require people to synchronise their actions . We have developed an experimental paradigm , the HoneyComb computer-based multi-client game , to empirically investigate human movement coordination and leadership . Using economic games as a model , we set monetary incentives to motivate players on a virtual playfield to reach goals via players' movements . We asked whether ( I ) humans coordinate their movements when information is limited to an individual group member's observation of adjacent group member motion , ( II ) whether an informed group minority can lead an uninformed group majority to the minority's goal , and if so , ( III ) how this minority exerts its influence . We showed that in a human group – on the basis of movement alone – a minority can successfully lead a majority . Minorities lead successfully when ( a ) their members choose similar initial steps towards their goal field and ( b ) they are among the first in the whole group to make a move . Using our approach , we empirically demonstrate that the rules of swarming behaviour apply to humans . Even complex human behaviour , such as leadership and directed group movement , follow simple rules that are based on visual perception of local movement .
Schools of fish and flocks of birds move collectively towards a spatial goal [1] , [2] despite their large local group sizes and therefore reduced capacity for global or inter-individual communication across the group [3] , [4] . Behavioural modelling [5]–[7] and empirical research [8]–[10] have shown that in diverse species , including humans [11] , [12] , local individual rules are adequate to generate complex collective behaviour at the group level [13]–[15] . There is increasing evidence [9] , [16] , [17] that not only large swarms but also small heterogeneous groups may be coordinated by local interaction rules . To explain this “swarming” phenomenon in animals and humans , Couzin et al . [18] created a model in which group locomotion emerges from individuals steering their motion based on the moves of local neighbours . Their model comprises three fundamental parameters described by Aoki [19] and Reynolds [20] , stating that members of a swarm ( a ) become attracted to neighbours' positions within a local range ( cohesion ) ; ( b ) align with neighbours' direction and speed within this range ( alignment ) and c ) avoid neighbours within a predefined radius ( collision avoidance ) . In order to incorporate the influence of those individuals with information about a preferred goal , a weighted direction vector was added [18] to investigate the dilemma of informed individuals pursuing their preferred goal while trying to remain with the group [21] . In their computer simulation model , Couzin et al . [18] deduced that a proportionally small number of directionally informed individuals can channel the naïve ( uninformed ) members of the swarm to the target of the directionally informed . Neither the informed nor the naïve need to recognise each other , be aware of the informational gap , or practise active signalling . Assumptions about inherent personal distinctions ( e . g . personality traits or social cues ) need not to be present in order to explain effective movement leadership . The purpose of the present study was to test whether such “swarm-like” human movement and leadership behaviour empirically holds for a small group of humans restricted to ‘reading/transmitting’ only movement behaviour . To do so , we developed the computer-based HoneyComb multi-client game as our investigative platform . The elements of this virtual game were designed to eliminate all sensory/communication channels except the perception of player-assigned avatar movements on the playfield . To create experimental factors of individual motivation towards the two swarming behaviours – “cohesion” and “alignment” – without experimenter's direct behavioural instructions , we implemented within-group graded monetary incentives , an informed minority of players with a higher-rewarded target , and an uninformed majority of players with lower but equal reward targets . This experimental paradigm differs substantially from the approach taken in three other studies on human movement and leadership by Dyer et al . [11] , [22] and Faria et al . [23] , where the minority leadership prediction of the Couzin et al . model [18] was investigated in a face-to-face group situation . All three studies conducted live experiments of humans walking in a circular arena with a landmark randomly assigned as a target to the informed individuals . This naturalistic approach has benefits in terms of ecological validity . But these studies fall short in discretising the spectrum of human communication: defining where communication among humans begins in the phenomenon of leading human group movement . Firstly , face-to-face procedures , despite instructions not to communicate , could not preclude the transmission of non-verbal cues between participants ( e . g . eye contact , facial expressions , shared/non-shared social categories of age and gender ) . These cues likely functioned to signal intention towards the goal or to unite participants to “flock” based on the perceived affiliation to the same social category , almost certainly confounding results . Second , instruction methodology in these studies was direct , which is incongruent with the above-mentioned swarming conditions as “local” rules . In the three studies' standard set of instructions , the model [18] parameter “group cohesion” was directly translated into the investigator's explicit instruction to participants to “remain as a group of eight ( ten resp . ) ” . “Collision avoidance” was implemented by the instruction “to stay within an arm's length of another individual” [11] . Our platform ensures internal validity because the study design of humans moving their avatars on a virtual playfield eliminates all sensory/communication channels between the human participants other than observed directional movement . By using monetary incentives we introduce a motivational factor for group movement but avoid any direct instruction; therefore , any potential leadership influence of movement remains with the transmitting/reading of participants' movements rather than external or internal socio-cognitive sources . The goals of our study are to address three fundamental questions regarding basic human coordination mechanisms and the emergence of patterns of group leadership in the complete absence of communication mechanisms and pre-knowledge of information differentiation other than the perceived movement of others: ( I ) Can humans coordinate their avatars' movement under such extremely restricted communication conditions ? ( IIa ) Can the informed minority lead the uninformed majority to their goal field , ( IIb ) even with the additional restriction for each group member to perceive only local movement relative to his/her proximity ? ( III ) Which movement behaviour of the minority is the best predictor for the success of leading the majority to a target ?
A total of 400 subjects participated in our experimental application of the HoneyComb game , conducted in a quiet zone of the main lecture building of a large university in Germany ( see Material and Methods for details on Pre-test , Participants , Experimental Procedure , Locomotion , and Ethics Statement ) . In the HoneyComb game , participants interacted anonymously in groups of ten players by means of computers connected via a local network . Study question ( I ) – can humans coordinate their avatars' movement under such extremely restricted communication conditions – required absolute exclusion of any sort of communication other than observed directional movement . Each player was surrounded by computer-station partitions and was required to use ear plugs to prevent verbal and nonverbal communication during the experiment . On the virtual playfield resembling a honey comb of 97 hexagonal spatial fields , each participant was represented as an avatar , i . e . as a black dot identifiable only to him/her by being twice the size of the other nine co-participants' dots ( Fig . 1 ) . At the beginning of the game , all players' dots were positioned in the centre of the honey comb . In each of the 15 available moves , the players could navigate their dot via mouse-click to one of six neighbouring fields from their respective point of departure . An incentive structure operationalising the model parameters was implemented via six spatial goal fields rendering monetary payoffs ( € or €€ ) ( alignment ) . If a player arrived at a payoff field , his or her payoff would be multiplied by the number of co-players' avatars standing on this payoff field at the end of the game ( cohesion ) . To gain a high payoff , coordinated choices were thus advantageous . The criterion for ending the game was either that all players' avatars stood on payoff fields , and/or that all players had used all available moves . To explore study question ( IIa ) where we tested the first experimental factor regarding the informed minority being able to lead the uninformed majority to their goal field , a minority/majority information differentiation within each of the ten-person groups was established: a minority of two randomly selected players was informed of the location of their one highly-rewarded €€-goal field in addition to five lower-rewarded €-fields ( Fig . 1a ) . A majority of eight players were notified of six equally lower-rewarded goal fields ( Fig . 1b ) . Neither the highly-rewarded “informed” nor the lower-rewarded “uninformed” knew whether they were in the majority or in the minority or that there was a reward difference among players . To address question ( IIb ) where we assessed whether locomotive coordination and leadership would be possible only within global or also within local perception radius of the players , the second experimental factor was implemented where members were additionally restricted to be able to perceive only local movement relative to his/her proximity . The 40 ten-person groups were randomly allotted to either the local condition ( n = 20 groups ) limiting the participants' sight to events only on the neighbouring fields of their dot's position or to the global condition ( n = 20 groups ) disclosing an overview of all events on the playfield . It is important to note that the higher incentive of the informed players potentially creating leaders does not necessarily generate follower behaviour; it simply establishes a motivation factor normally present in groups where leaders and followers emerge . Both behaviours – leading and following – are necessary for effective leadership to occur [24] . That the majority , i . e . a higher than probabilistically-expected number of non-informed group members , will follow said informed individuals is not guaranteed by the mere presence of a motivation factor as it is yet to be seen whether its presence invokes any sort of group-member leading and following behaviour . If leadership of the informed minority is indeed observed , it then piques the question of which characteristic ( s ) of their movement behaviour makes those minority members successful in leading the uninformed majority members to their goal field . In the choice of variables characterising the informed players' movement behaviour most apt to successfully lead the majority to the minority's preferred field ( question III ) , we focused on the initial stage of the game , as these behavioural variables ( effect of local-only or globally-perceived movement by the minority ( informed ) and majority ( uniformed ) members; whether and to what extent being the first minority leader to move has an effect on follower behaviour; and effect of path consistency among the minority leaders ) have a greater claim for being influential and thus allowing for a ( tentative ) causal interpretation of our results . This section describes the variables coded to test our hypotheses regarding the relationship between the visibility radius of the players , the initial moves of the informed players and their leadership success , which is measured by the number of arrivals of uninformed players on the informed players' preferred outcome field . For all players , only the money depots had a subjective value . We therefore expected that 0% players would arrive elsewhere in the hexagon . For the uninformed majority , the one ‘€€’ money depot ( unknown to them ) was expected to be as attractive as each of the five ‘€’ money depots , leading to a predicted probability of a sixth ( 16 . 67%; n = 53 ) of the uninformed majority to arrive at the ‘€€’ money depot by the end of the game . However , empirically , 34 . 37% ( n = 110 ) of uninformed players arrived at the ‘€€’ money depot . In order to test the null hypothesis that the arrival rate on the ‘€€’ money depot was less or equal to a sixth , we implemented a two-stage bootstrap procedure that accounted for the hierarchical structure of the data: 1 , 000 bootstrap estimates of the arrival rate were obtained by first sampling 40 groups with replacement before sampling eight players within each group with replacement . Our result indicated that the arrival rate was significantly higher than a sixth ( P<0 . 001 ) . This means that informed minority players , even in the absence of verbal and non-verbal communication , were able to significantly influence their uninformed majority co-players to move their avatars to their ‘€€’ target , primarily because informed minority players initiated movement earlier than uninformed majority players: in 72 . 5% of all observed groups , informed minority players showed a mean starting rank less than the middle starting rank of 5 . 50 for their initial move ( binomial test: P< . 001 ) . Fig . 2 shows a histogram of the number of arrivals of non-informed players' avatars on the informed players' preferred outcome field ( arrivals ) for the 40 groups ( grey bars ) . In principal , arrivals might take on all discrete values between zero and eight . Hence , a starting point for modelling the distribution of the variable arrivals would have been a binomial distribution with eight trials . However , from visual as well as statistical inspection of the univariate distribution , we stipulated a finite mixture model entailing two binomials with success probabilities π1 and π2 and mixing parameter α . Parameters were estimated via maximum likelihood methods ( ) . The probability mass function of this mixture model and the contributions of the two components are depicted in Fig . 2 ( dashed lines for the first component , dotted lines for the second ) . We successfully tested the null of one component against the alternative of two components following the parametric bootstrap procedure proposed by McLachlan [33] ( P< . 001 ) – in this case , by calculating the likelihood-ratio test statistic for 200 bootstrap samples from the null model . In a second step , we introduced covariates to the model by letting the mixture parameter α depend on a linear combination of variables describing the initial movement behaviour of the informed minority . Descriptive statistics for all variables and associated hypotheses are summarised in Table 1 . Negative coefficients signal positive associations between arrivals and the respective variables in the model we specify , as we explain in the next section . As mentioned above , covariates affect the outcome variable through the mixing distribution . Hence , the distribution of , the number of arrivals in game i , given a vector of covariates , is calculated bywhere and the s are parameters to be estimated; is the probability mass function of a binomial distribution with eight trials; is the logit function; and , the matrix of covariates , includes a constant . Note that for ( which happens to be the case in our application ) , a positive parameter value signals a negative association between the number of arrivals and the variable of interest . Results of our statistical analyses are summarised in Table 2 . In addition to parameter estimates and asymptotic z-values for the s , we report the number of parameters ( p ) , the retained minimum of the negative of the log-likelihood ( -l ) , and the Akaike and Bayes model selection criteria ( AIC and BIC , respectively ) . The first column reports results from an “empty” model , i . e . the model without any covariates described above where only the mixing parameter is transformed to a constant via the logit link . Despite being non-informative in terms of the determinants of the number of arrivals , this three-parameter model was the most parsimonious and already yielded a reasonable fit . It therefore served as a useful comparison . As described above , we also measured different variables describing the overall locomotive behaviour of the informed players . Incorporating these variables in our empirical model , we found that , while sometimes significant , these specifications generally resulted in inferior model fits . Informed decisions taken during the very early stage of the game led to very distinct outcomes , as confirmed by the distinctly and visibly bimodal distribution of the dependent variable “arrivals of the uninformed players” ( see Fig . 2 ) . The second column reports results from a model variant in which we tested for the effect of the local-condition . We found a mildly negative coefficient on the local-dummy . That means that constraining the perception of the players tended to increase the number of arrivals on the informed players' preferred field . However , the coefficient is insignificant at conventional levels . In column three we tested whether the success probability increased when one of the informed players was the first to move . The coefficient does have the expected negative sign but is also insignificant . To further investigate these tendencies , we interacted the variable first with local . Demonstrating decisiveness by making the first move might have been more important when there was a real threat of losing sight of other players . This is indeed what we found , as demonstrated by the significant coefficients in column four on first and the first×local interaction term . In column five , we tested whether moving into the same direction increased the success probability . We found that both variables , same and direction , exerted a positive effect on the number of arrivals , with a quantitatively larger effect for the former . We conclude that a condition of informed minority players moving in cohesive lockstep strongly predicts the outcome of the game . In addition , the model variant reported in column five was the only model variant that performed superior in terms of both model selection criteria when compared to the non-informative model variant in column one . Fig . 3 and 4 depict histograms of the arrivals under conditions specified in columns 4 and 5 of Table 2 , respectively , against the fitted model variants . Making the first move was associated with a lower number of arrivals if players disposed of a global overview of the playfield . Only in 4 out of 20 games was the player able to make the first move as one of the informed . If the limiting local-condition view of adjacent players only was active , the predictor first was associated with an increasing success probability , as can be seen from the lower two panels in Fig . 3 . The fit of the model variant in column 5 is demonstrated in Fig . 4 . Note that in the eight games in which the informed players did not move their avatars into the direction of the preferred field , the number of arrivals was always zero . Hence , the second distribution received almost zero weight . This predicted number of arrivals changed drastically when the two informed minority players demonstrated cohesion and thus provided other players with a signal about their preferred direction .
We confirmed that even in humans , self-organised patterns of collective movement emerge without assuming global information , strategic or other cognitive complex considerations , or any other sort of communication besides transmitting/reading one's own and other participants' movement behaviour . We could predict the most effective locomotive behaviour of an informed minority to pull an uninformed majority's avatar movements to a specific goal field ( i . e . invoking effective leadership ) when all channels of communication were reduced to pure movement on a virtual playfield , even when perception radius limited transmitting/reading movement information to/from local neighbours [3] , [4] , [34] . The successful locomotive pattern of the informed minority players was achieved by moving their avatars analogously into the same direction , and – when perception radius was restricted to only neighbouring fields – making use of an initial mover's effect [28] . The results show that the immediacy of the very first move of the informed minority players and consistency of their initial move – compared to their overall locomotive pattern , path similarity , path length , movement latency , and starting order – are the most effective behaviours for influencing the majority to follow . Personality traits and computer literacy of informed minority players were not crucial to their success in pulling the uninformed majority's avatars to their preferred goal field . Our results corroborate one of the core assumptions of Couzin et al . 's [18] modelling of the synoptic process of individuals coordinating via movement as a channel of information transfer . This empirical evidence that the ability exists to coordinate behaviour via the transmitting/reading of movement alone might have high adaptive value in situations where human groups experience restricted communication and therefore are forced to lead solely via movement trends of immediacy and consistency . Applied to emergency , rescue , and sport scenarios where face-to-face communication is hindered but movement is still possible , the group success rate towards a desired goal could be maximised via movement initiation that is decisive: i . e . consistent , immediate and therefore consequent . For instance , leadership personnel of such groups could be trained accordingly in using simple behavioural mechanisms for leading masses of uninformed people to emergency exits or secure areas . But the question here is whether our results – obtained in a virtual environment where movement was performed by representations of the group participants as black dots on a virtual playfield – are applicable to these actual ( vs . avatar ) human movement scenarios . It can be argued that the representation of the human participants in the rather abstract avatars of black dots in a virtual environment exacerbated the lack of sense of connection between the participants [35] . Because our avatars were designed to reduce social meanings and social roles to a minimum , coupled with the work by others showing avatars mediate the body in virtual settings [36] and that anthropomorphic as well as polymorphic representations in a virtual environment facilitate feelings of embodiment [37] , we felt confident that human behavioural association would not be diminished by not transmitting influences of body shape , gender , age and other hierarchical standings [38] by our HoneyComb avatars . Also , the physical presence of their co-players sitting at tables besides them – although behind partitions and with earplugs – likely confirmed feelings of human embodiment . Although we did not ask participants whether they felt sufficiently represented by their avatar and also perceived their co-players as humans moving on a field in this study , the high percentage of participants reporting high levels of embodiment in a subsequent , yet to be published study might hold as an additional argument . The use of movement as a basic signal to maintain group cohesion and indicate direction appears to be an innate behaviour that does not require complex cognition [39] , [40] . As in models describing collective pedestrian behaviour as “spatiotemporal patterns” emergent “through the nonlinear interactions of pedestrians” [12 , p . 368] , we could empirically show that collective group movement and leadership – in other words non-random behaviour – emerged empirically from implementing the assumed parameters of the swarm behaviour models [18]–[20] into incentives within our HoneyComb virtual movement game . Unlike studies on pedestrian behaviour [16] , [17] , [41] , we did not base our approach on the sociological concept of a group allowing social interaction and social ties between individuals . Nevertheless , fundamental human mechanisms of collective movement as trails , unidirectional lanes or a basic principle of least effort [7] identified by these researchers are likely to underlie the movement patterns identified in our study . To explain such complex collective phenomena in humans , we needed neither to assume humans communicating with each other , nor to apply other higher order cognitive and/or social competence nor mutual acknowledging of intentional behaviour in leading ( and being led ) as none of the players knew that there was any informational difference in the group . Methodologically , this means that functional behavioural complexity at the group level does not necessarily equate with an underlying cognitive complexity at the individual level , but can also be explained by distributed embodied cognition [39] , local heuristics [42] or even the principle of least effort [7] . Due to the physical restrictions of our study setup , as compared to the Dyer et al . experiment setting [11] , [22] , [23] , it is highly unlikely that players were able to signal each other their intention , cognitive or motivational state . The rationale behind our study's strident physical restrictions was to focus on mere movement behaviour as a type of inadvertent social cue or – even less – the possibility that the players do not apply mutual responsiveness but in the simplest case practise spatial pattern recognition [39] , suggesting the most reductionist explanation possible for effective leadership behaviour in human groups . Methodologically , this study's application of the HoneyComb paradigm makes heuristic use of existing formal models of swarming behaviour [18]–[20] in order to ( a ) implement the model parameters into experimentally set behavioural incentives to participants , ( b ) test whether their empirical behaviour fits specific model predictions in this study of leadership of a minority over a majority , and ( c ) describe patterns of this observed behaviour . With step ( c ) we reach beyond the existing formal models we built upon . The next investigations will likely entail additional applications of the HoneyComb computer-based multi-client game approach to analyse the process patterns on a more sophisticated level , i . e . Markov chains and topological models , and design and execute further experiments in order to identify additional influencing factors of leadership ( e . g . initial positioning , visual appearance , communication , and the identifiability of informed minorities ) . Along the lines of the experimental studies on schooling fish by Couzin et al . [43] and Tunstrøm et al . [44] , we could gain deeper insights into the mutual dependence of leadership and followership by manipulating size as well as informational status of the subpopulations of a group and of the size of the group as a whole . We have already run further HoneyComb paradigm based experiments where we manipulated the colours of the avatars in order to investigate how the distinction of “minimal groups” [45] of similar individuals can influence their flocking behaviour and their mutual perception as well as their identification with “their” group . Another application of the paradigm has been to install two minorities with opposing goal fields to investigate how leadership and group fission and fusion work under these conditions . In sum , our approach and the results of this study provide a new paradigm on boundaries of communication in the influence of coordinated human movement that could readily be extended to additional questions regarding consensus and leadership dynamics .
Data collection and data analysis procedures in this project “Leadership in coordination games” have been approved by the Ethics Committee of the Georg-Elias-Müller Institute for Psychology of the University of Göttingen ( proposal 039/2012 ) . The impact of cohesion incentives independent of other factors was tested in five ten-person groups . All groups consisted only of uninformed players with six equal reward targets ( ‘€’ money depots ) . The five groups reached a mean arrival rate of 92% on one single ‘€’ money depot at a time ( s . d . = 8 . 37% , minimum of 80% , maximum of 100% ) .
|
Our article gives empirical evidence of group coordination mechanisms and basic rules of leadership that assist in leading a human group . Using a computer-based multi-client game that blocks explicit signals or other typical human information transfer , we offer a model of human group movement patterns applicable to group scenarios such as emergency , rescue , and sports where inter-individual communication is hindered but the reading of movement is still possible . Results show that even in these communication-restricted situations , movement of an informed minority that is efficient and consistent can effectively pull the majority towards a target goal .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"psychology",
"psychological",
"adjustment",
"behavior",
"biology",
"and",
"life",
"sciences",
"social",
"psychology",
"social",
"sciences",
"evolutionary",
"biology",
"evolutionary",
"emergence",
"evolutionary",
"processes"
] |
2014
|
Leadership in Moving Human Groups
|
Incentive salience is a motivational property with ‘magnet-like’ qualities . When attributed to reward-predicting stimuli ( cues ) , incentive salience triggers a pulse of ‘wanting’ and an individual is pulled toward the cues and reward . A key computational question is how incentive salience is generated during a cue re-encounter , which combines both learning and the state of limbic brain mechanisms . Learning processes , such as temporal-difference models , provide one way for stimuli to acquire cached predictive values of rewards . However , empirical data show that subsequent incentive values are also modulated on the fly by dynamic fluctuation in physiological states , altering cached values in ways requiring additional motivation mechanisms . Dynamic modulation of incentive salience for a Pavlovian conditioned stimulus ( CS or cue ) occurs during certain states , without necessarily requiring ( re ) learning about the cue . In some cases , dynamic modulation of cue value occurs during states that are quite novel , never having been experienced before , and even prior to experience of the associated unconditioned reward in the new state . Such cases can include novel drug-induced mesolimbic activation and addictive incentive-sensitization , as well as natural appetite states such as salt appetite . Dynamic enhancement specifically raises the incentive salience of an appropriate CS , without necessarily changing that of other CSs . Here we suggest a new computational model that modulates incentive salience by integrating changing physiological states with prior learning . We support the model with behavioral and neurobiological data from empirical tests that demonstrate dynamic elevations in cue-triggered motivation ( involving natural salt appetite , and drug-induced intoxication and sensitization ) . Our data call for a dynamic model of incentive salience , such as presented here . Computational models can adequately capture fluctuations in cue-triggered ‘wanting’ only by incorporating modulation of previously learned values by natural appetite and addiction-related states .
Incentive salience is essentially a Pavlovian motivational response: it takes Pavlovian associations as its learned input . Potential differences between cached valence and motivation values for Pavlovian cues have been previously noted by computational modelers and learning theorists [14] , [18]–[22] . For example , Dayan and Balleine noted that , “Pavlovian CRs are … directly sensitive to the level of motivation of the subjects , such as whether or not they are hungry . This motivational or affective control argues against simple realizations of the critic in which there is one set of weights or parameters ( whatever their neural realization ) mapping from stimulus representations ( i . e . , units whose activities are determined by the stimuli shown to the subject ) to the values” ( p . 288 , [14] ) . Here we aim to account for such motivation variations by identifying incentive salience computations that dynamically determine Pavlovian motivation value . These computations must integrate Pavlovian learned inputs with the status of the brain mesolimbic mechanisms that reflect current physiological states ( hungers , drug intoxication , sensitization , etc . ) . To see the difference between the vantage points of incentive salience compared to reward learning , it may be helpful to review how cached values are established in contemporary reinforcement learning models . For example , the temporal-difference ( TD ) method provides an explicit “model-free” formula for estimating expected reward . It is model-free in the sense that it does not involve an internal model or cognitive map of the world , but depends only on cached experiences and the accumulated state for estimating the value function . The estimate is based on summed values from reward prediction errors – the discrepancy between the reward expected from a stimulus ( technically , a state ) and the reward actually received [20] , [23]–[29] . One influential neurobiological view identifies the predictive error signal , which lies at the core of temporal difference learning , with the firing of dopaminergic neurons projecting to the nucleus accumbens and neostriatum [30] , [31] ( though this notion is not without controversy regarding the causal role of dopamine systems in generating prediction errors and value estimates [2] , [32] ) . The actor-critic architecture , along with the TD-based learning rule , carries great computational power . It provides , at least theoretically , a consistent and effective scheme to solve the so-called “dynamic programming” problem [33] concerning optimization for sequential decision-making under a stationary Markov environment , without the need of an elaborate model of the world ( i . e . , how states unfold successively and how one's actions affect such state-transition ) [27] , [31] , [34]–[36] . The TD error model also has been applied to incentive salience [26] . A first step was taken by McClure , Daw and Montague , who used the concept of prediction error-driven learning , and equated cue-triggered ‘wanting’ with the cached Pavlovian learned value acquired by the TD method [26] . Their model postulated incentive salience to function as a cached , cumulative reward value , which , if a reward was suddenly revalued , could be changed only by further relearning about a new prediction error by re-encountering the UCS . They suggested that “the role of dopamine in learning to attribute such expectations to situations that are predictive of reward and in biasing action selection towards such situations as the formal counterpart to the ideas of Berridge and Robinson about the role of dopamine in attributing and using incentive salience . ” p . 425 , [26] . While a useful contribution that exploited the TD model's strengths to capture trial-by-trial ‘reboosting’ of ‘wanting’ , we believe it is only an initial step towards modeling incentive salience . Despite the computational success of actor-critic architecture , several theorists have suggested , as noted above and consistent with our view here , that additional mechanisms must be added to explain emerging psychological and neurobiological data and to account for motivation because the simple form of actor-critic architecture produces a rigidly incremental cached value of prediction [14] , [37] , [38] . One solution to account for immediate state-based changes in behavior has been to posit that dopamine also modulates the “vigor” of all responses in a general fashion [37] , [38] . Additionally , uncertainty or generalization decrements after a state change has been suggested to explain some reduction in behavioral responses , at least for changes that devalue a reward downwards [39] . Another alternative solution is to sidestep the model-free cached limitations and add an entirely separate model-based ( e . g . , cortically-embedded ) mechanism for reward prediction in the form of a model-based or tree-search system that explicitly represents the world as a cognitive “state space” . Such a cognitive model includes goal values and act-outcome ( A-O ) relationships , and which can adjust instrumental behavior more flexibly , at least once a new goal value is known by experience [14] , [17] , [18] , [21] , [22] , [39]–[41] . Still , cue-triggered ‘wanting’ differs from all the above , and we believe may more accurately capture the chief motivation function of brain mesocorticolimbic systems . The re-computations of the incentive salience for a Pavlovian CS may in some cases be carried out in a highly dynamic , stimulus-specific and stimulus-bound fashion , as will be described below . This is possible because of Bindra-Toates rules of Pavlovian motivation that underlie incentive salience [3]–[5] . Those rules are distinct from both cached TD values and model-based cognitive predictions . Cached values and model-based predictions of reward both often assume that a reward will be about as good in the future as it was in the past [14] , [22] , [26] , [40] . Robust computational theories exist for those cached model-free values and cognitive model-based systems , but not yet for Bindra-Toates computations of incentive salience . A similar sentiment was recently expressed by Dayan and Niv , “Unfortunately , the sophisticated behavioral and neural analyses of model-free and model-based instrumental values are not paralleled , as of yet , by an equivalently worked-out theory for the construction of Pavlovian values . ” ( p . 191 ) [22] . Our goal here is to take a small step towards a better computational theory for Pavlovian-guided generation of incentive salience . Specifically , we aim for a model able to compute cue-triggered ‘wanting’ even for novel physiological modulations that occur before there is an opportunity for relearning the altered reward . We also aim for a model that can account for ‘irrational wanting’ in addictions; that is , for excessively ‘wanting’ a reward even when knowing its future value to not deserve intense motivation . Consistent with contemporary views , we believe that multiple reward-related learning processes exist within a single brain , mediated by separable brain systems [7] , [14] , [17] , [18] , [21] , [22] , [38]–[40] , [42]–[45] . We presume these reward learning-motivation mechanisms occur in parallel as three separable processes ( S-S Pavlovian-guided incentive salience , S-R cached habits , and model-based cognitive expectations ) . Our model is meant to capture only incentive salience transformations , which takes Pavlovian CS-UCS associations involving rewards as the primary learned input for Bindra-Toates modulations . Our view of incentive salience calls for the dynamic computation of “incentive value” of a conditioned or unconditioned stimulus , where ( a ) the CS stimulus has previously been associated with the relevant UCS; and ( b ) the value is gain-controlled moment to moment by fluctuations in relevant physiological states ( including neurobiological states of brain mesocorticolimbic systems ) . Recall that in reinforcement learning , the expected total future discounted reward ( or simply average reward value ) V associated with a state s ( i . e . , the conditioned stimulus ) is ( 1 ) where is the discount factor , rt , rt+1 , rt+2 … , representing the sequence of primary rewards starting from the current state ( subscripted t ) , and the expectation 〈·〉 is taken over possible randomness in environmental state transition and reward delivery ( the bracket sign around primary reward values will be omitted below for clarity ) . The estimated value of reward prediction ( denoted with a hat ) is a value gradually acquired by the agent through temporal difference learning over a series of experiences in which the predictive CS and UCS reward are paired . The acquisition of reward estimate is based on computing a prediction error δ correcting any experienced deviation from consistent successive predictions: ( 2 ) and then updating according to . The value function V is essentially an incrementally-learned associative prediction of each state . As mentioned , one previous computational proposal for incentive salience equated this value function V , defined by Eqn ( 1 ) , with the motivational concept of CS incentive salience [26] , and gradually altered motivational value by increments in V in each trial produced by prediction error at the moment when reward UCS was received , via the temporal difference error variable δ , defined by Eqn ( 2 ) . The TD error δ was identified with the “reboosting” process of a CS posited by the incentive salience hypothesis to occur at the moment of UCS [1] , [46] . Reboosting was a concept added a decade earlier by the original incentive salience proposals to account for the gradual decrement effects on rewarded behavior produced by administering neuroleptic drugs that partially blocked dopamine receptors ( anhedonia-like effects or extinction mimicry ) [47] . In such a conceptualization , the incentive salience of a stimulus is essentially the accumulated reinforcement value of such a conditioned stimulus acquired through TD prediction-error learning .
All animal work was conducted according to relevant University of Michigan , NIH , national and international guidelines . We suggest additionally that a model of incentive salience should also incorporate dynamic physiological modulation by current states to capture more sudden changes in CS motivational value , such as in specific appetites or drug enhancement of ‘wanting’ , which do not proceed by gradual reboosting via UCS prediction errors [1] , [2] , [48] , [49] . In these conditions , a CS's motivationally-transformed incentive salience value may dramatically diverge from cache-generated predictions of reward [10] , [12]–[14] , [40] , [44] , [50] . In at least some cases these physiological revisions of value can occur without any need of relearning about the change in UCS hedonic impact to revise CS-UCS predictions , yet still be so powerful as to completely reverse the incentive value of a CS . In the following , we will first propose a model for incentive salience that can incorporate dynamic modulation of cue-triggered ‘wanting’ by even novel physiological states . Next , we will show how such a model maps onto empirical evidence for dynamic modulation in examples of natural appetite , amphetamine intoxication , and addiction-related sensitization . To describe our model of incentive salience more precisely , here we propose that , respecting the difference between motivation and learning , incentive salience computations incorporate a physiological factor κ that modulates the value of a CS associated with a relevant UCS ( which carries a reward value of rt ) . The κ factor reflects current physiological state ( hungers , satieties , drug states , etc . ) . The role of κ is to allow incentive salience of an associated CS to be dynamically modulated by physiological factors relevant to future rewards ( e . g . , hungers , satiety , drug intoxication , mesolimbic sensitization , etc . ) . We suggest that the incentive salience or motivational value of the current state in the presence of a reward CS is ( 3 ) Equation ( 3 ) represents a generic model for incorporating the motivation factor κ which modulates the learned representation of primary reward and hence the notion of incentive salience , where κ is a positive constant that varies with behavioral state . Here is a generic two variable function which , in the following discussions , will be specialized to either of two forms ( sub-types ) , as described below . ( 3a ) ( 3b ) Below , we refer to κ as the “gating parameter” for incentive computations involving physiological manipulations for a previously learned CS ( e . g . hunger , thirst , salt appetite , etc . ) , with κ<1 representing devaluation ( such as satiation ) , and κ>1 representing enhancement ( such as appetite or sensitization ) . When κ = 1 our model reduces to the conventional temporal difference model; that is , when physiological state remains constant across training and test . When physiological state changes from training to test , one of the two special versions of Equation 3 will apply , and which of the two is most appropriate will depend on the situation ( Figure 1 ) . Equation ( 3a ) describes a specific subtype in the form of a multiplicative mechanism , appropriate for most situations where motivation changes from low to high or vice versa without changing valence — reward is manipulated between 0 and a positive value , changing incentive salience from neutral to ‘wanted’ ( or from ‘wanted’ to neutral ) . In these cases , κ is a gain-control factor that scales up ( i . e . , magnifies , when κ>1 ) or down ( i . e . , shrinks , when κ<1 ) the incentive salience of the reward . Equation ( 3b ) describes another subtype in the form of an additive mechanism , appropriate for other situations where incentive value not only changes but reverses valence between positive and negative — this can additionally account for any special cases in which a reward value changes polarity from positive to negative or from negative to positive . In those cases , the logκ term moves the baseline of the incentive salience value , which can be shifted either up ( i . e . , increases ‘wanting’ , κ>1 ) or down ( i . e . , decrease ‘wanting’ , κ<1 ) . This allows polarity reversal from a negative value to a positive value ( with κ much larger than 1 ) , or vice versa ( with κ closer to 0 ) . The reason why we include both an additive and a multiplicative version of modulation in Equation ( 3 ) is to more sensibly achieve real-life reversals than can be accomplished in a purely multiplicative model by simply changing the κ valence polarity to negative . This is because merely changing polarity in a multiplicative Equation ( 3a ) would invert the rank order when multiple reward stimuli in the same family were involved ( e . g . 3 concentrations of salt ) . That would revalue the respective order of the series in ways that might be unrealistic . For example , reversing valence in a multiplicative model would cause the reward that was originally most highly liked and ‘wanted’ of all to become the most highly disliked or repulsive after devaluation of all; an intermediately liked reward would become intermediately disliked , while a nearly neutral reward stimulus would remain nearly neutral after polarity reversal . Such re-ordering fails to describe what happens in empirical cases of valence reversal , where the originally most liked reward may often still remain the best of a bad lot , becoming the least disliked as a physiological manipulation changes the valence of the entire group . By contrast , an additive model as expressed in Equation ( 3b ) , allows the ‘best’ stimulus to remain best relative to the others , even if their absolute values may switch valence ( i . e . , all shift across zero ) . Specific candidates for polarity reversal include reversals in reward values from nasty to nice , such as described below where an intensely salty taste reward is re-encountered during a salt appetite , or from nice to nasty such as after taste-aversion devaluation [10] , [50]–[52] ( Figure 1 ) . Polarity reversal would similarly encompass cases in which motivational salience changes valence between desire and dread [53] , [54] . We remark in passing that we use logκ instead of κ for the additive Eqn ( 3b ) simply to have the same parametric representation in the additive case as the multiplicative case . Also note that we only consider additive and multiplicative mechanisms which together generate the group of ( positive ) affine transformations on reward values ( this is the class of transformation that keeps the optimal policy invariant [55] ) . From Eqns ( 3 ) and ( 1 ) , and assuming a multiplicative mechanism , the incentive value is related to the average reward V ( i . e . , total reward including current and all future-discounted rewards ) via ( 4 ) In essence , Eqn . ( 4 ) is an expression of what is known as the “quasi-hyperbolic” discounting model encountered in economics literature [56] , [57] . The proposal here exploits the two parameters in the quasi-hyperbolic discount model format and is the basis of our current postulate of a gain-control mechanism to implement incentive salience computations , in a manner that can be sensitive to inter-temporal comparisons of values when such comparisons play important roles [58] , cf . [59] . In simple terms , our model of incentive salience reduces to in the absence of devaluation/sensitization manipulation ( κ = 1 ) . The modulatory factor κ is assumed to be independently controlled by the geometric temporal discounting under γ , though it is possible that such changes can be coupled . For example , sensitization or an increased physiological appetitive state ( κ becomes greater than 1 ) might lead to a decrease in the temporal horizon γ [60] , producing sharper temporal discounting effects , such that motivational value increases with degree of temporal proximity to reward UCS [61] . Note that the incentive value of a state st is the motivationally-modulated value of the immediate reward rt plus the discounted value of the expected reward in the next state st+1; both these are loaded into the goal representation as st is presented . The incentive salience hypothesis specifically proposes that Pavlovian-guided attribution of incentive salience is mediated principally via subcortically-weighted NAcc-related circuitry involving dopamine neurotransmission , which pass signals through the ventral pallidum . These circuits include input from mesolimbic dopamine projections from ventral tegmentum and substantia nigra to the nucleus accumbens , ventral pallidum , and amygdala; and output projections from nucleus accumbens that converge through ventral pallidum [62]–[64] . From ventral pallidum these signals then pass to a thalamic relay for return to mesocorticolimbic loops , or directly descend to other subcortical outputs [62] , [65] , [66] . In addition to receiving mesolimbic outputs , dopamine projections from VTA also ascend directly to ventral pallidum [67] , [68] . Thus the incentive salience hypothesis views incentive salience or ‘wanting’ to be influenced by dopamine-related modulations of function within this circuit , the output of which passes through ventral pallidum as a limbic ‘final common path’ . The computational approach suggested here can therefore be tested empirically by measuring neural signals carrying incentive salience in the final common path through ventral pallidum , in experiments which manipulate NAcc-related circuitry via changes in natural appetite states ( hungers , satieties ) or via addictive drugs ( drug administration; long-term drug sensitization ) .
To illustrate our proposal about the computation of incentive salience , we now draw on two types of experiments designed to expose dynamic physiological modulation of cue-triggered ‘wanting’ , as posited in equation ( 3 ) . The first experiment uses the natural motivation state of salt appetite to change the incentive salience of a salt CS . The second experiment uses a dopamine-stimulating drug amphetamine and/or enduring drug-induced sensitization to activate mesolimbic NAcc-related systems and change the incentive salience of a sucrose CS . A special case of incentive salience modulation is incentive-sensitization: this occurs when drugs in the brain sensitize mesolimbic dopamine-related systems , and similar but temporary elevation of ‘wanting’ can be produced by directly injecting amphetamine before a test [8] , [13] , [72]–[74] . We capitalized on these drug-induced elevations in incentive salience to test the addiction-related predictions of ( Eqn 3a ) for enhancing cue-triggered wanting’ [8] . To compare these neuronal coding formulations against our model for incentive salience , we developed an analytic technique called Profile Analysis to assess neuronal responses to CS1 , CS2 and UCS [13] , [76] . Profile Analysis creates a quantitative index comparing the ordering of the magnitudes of a neuron's firing rates to the three stimuli , CS1 , CS2 , and UCS ( Figure 3 ) . The profile for each unit is defined as a vector in a two dimensional “profile space” . The direction of this vector reflects the rank-ordering of each neuron's firing rate responses to CS1 , CS2 and UCS , while the magnitude of the vector reflects the degree to which the intensity of response to one stimulus dominates the responses to others ) . Inhibition of neuronal firing to a particular stimulus pulls coding vectors in opposite direction from excitations . All possible firing profiles are represented on a continuum of circular scale ( 360° ) , with nearby directions ( angles ) representing similar neuronal firing profiles . The profile analysis is performed on each individual neuron , and subsequently aggregated to obtain the entire neuronal population response . More formally , let us denote each neuron's firing rate to CS1 , CS2 , and UCS ( after normalizing to baseline ) as x , y , z respectively [77] . The relative rank-ordering of these three numbers according to their magnitude represents the “profile” of a neuron's responses to the stimuli , and it can be represented mathematically as a vector in a two dimensional space . For each neuron we construct a two-dimensional unit vector ( u , ν ) from the three numbers x , y , z , such that they ( i . e . , the profile-representing unit vectors ) are “equally spaced” in the projected two-dimensional subspace orthogonal to the direction [1 , 1 , 1]: ( 6 ) where the “anchoring” parameter α can be chosen arbitrarily . For simplicity , we chose α = 0; in this case , ( 7 ) The components of the profile vector [u , v] thus computed , according to Eqn . ( 6 ) in general and Eqn . ( 7 ) in particular , capture the two orthogonal contrasts formed among the three dependent variables x , y , z , such that any other contrast is a rotation in the two-dimensional space . This vector's magnitude ( 8 ) represents the extent to which the neuron's firing rates , x , y , z , are differentially modulated by the three types of stimuli ( CS1 , CS2 , and UCS ) , or in other words , it represents the variance of responses across the stimuli . The vector's direction ( 9 ) reflects the type of rank-ordering of the magnitudes of these firing rates . This procedure is both exhaustive ( i . e . , all neurons can be characterized ) and faithful ( i . e . , the distance between the angles is monotonically associated with the magnitude of difference in two profiles ) . Of particular interest to this study are the regions corresponding to response dominance by a particular stimulus ( Figure 3 ) . The region spanning 60° to 180° is where CS1 dominates the response profile and represents neurons that are responsive to the CS1 cue which carries the most predictive information about subsequent stimuli . This is designated as the prediction or “TD error-coding” area . The region spanning −60° to 60° represents dominant neural firing to CS2 , which occurs at moment of highest motivation excitement , and we denote it as the motivational or “incentive-coding” area . Finally , the region spanning 180° to 300° represents dominance by the reward itself ( UCS ) and it is designated the hedonic or “value-coding” area . Strictly speaking , our incentive salience theory predicts CS2>CS1>UCS for incentive-coding neurons , whereas if neurons obey a TD learning model , one predicts that the relative ordering of the magnitude of responses to the three stimuli after learning is CS1>CS2>UCS for error-coding neurons and UCS>CS2>CS1 for value-coding neurons . The rationale is similar to a method proposed by one of us earlier ( called “Locus Analysis” ) to characterize neurons in the primary motor cortex [78] . The results of the amphetamine and sensitization experiment revealed that VP neurons ordinarily signalled best the prediction value of a CS , responding maximally to CS1 ( Figure 3 ) . Thus , in the normal state , these limbic circuits reflect the standard prediction error model . However , mesolimbic activation or sensitization changed this profile by enhancing only incentive salience signals to the CS2 , at the expense of the signal for CS1 ( and without altering UCS signal ) [13] ( Figure 4 ) . The incentive shift toward CS2 was even greater for the combination of sensitization plus amphetamine administration at the time of test . The effects of the various mesolimbic dopaminergic activations can be visualized as the rotation of the Population Profile Vectors ( Figure 4 ) . Thus , it was concluded that while VP neurons in control animals ( after training ) tend to follow a TD error coding profile , mesolimbic dopaminergic activation causes the neuronal response profiles to shift towards encoding incentive salience . This shift corresponds to our motivational transform computation model , and to the idea that mesolimbic stimulations enhanced κ . Crucially , for showing the dynamic nature of the incentive increase , we note that the enhancements of neural firing to CS2 produced by amphetamine and by drug sensitization were evident right away on the very first presentations of the CS2 in the new sensitization and/or amphetamine state . That is , as predicted by our κ model , the incentive value of the CS2 was dynamically increased without need of re-learning about CS-UCS association , and without additional pairings with the UCS in the transformed state [13] . Does amphetamine or sensitization of incentive salience translate into behavioral ‘wanting’ too ? In previous studies using a rigorous behavioral test of cue-triggered ‘wanting’ ( based on a Pavlovian-Instrumental Transfer design or PIT ) , we and others have confirmed that acute amphetamine administration and/or prior drug sensitization both enhance peaks of cue-triggered ‘wanting’ for sucrose reward ( Figure 4 ) . In PIT , the phasic peaks of cue-triggered ‘wanting’ are manifest as a burst of pressing by the rat on a lever that previously earned sucrose reward: these peaks were dynamically enhanced by microinjection of amphetamine directly into the nucleus accumbens , or by sensitizing drug binges given weeks earlier [72] , [73] ( Figure 4 ) . The ‘wanting’ enhancements occurred even on the first presentations of the CS in the new physiological states of mesolimbic activation , just as in the neural firing experiments above ( Figure 4 ) [72] , [73] . And the elevations came and went with the coming and going of the physical CS+ stimulus , which lasted about 30 sec each . Such dynamic enhancement of CS incentive salience is also consistent with other behavioral demonstrations of incentive motivation enhancement by pharmacological dopamine activation or by psychostimulant-induced neural sensitization [72] , [73] , [79] , [80] , even in the absence of additional learning [81]–[83] . Our conclusion is also compatible with other behavioral evidence that the most predictive CS can be dissociated from the most ‘wanted’ CS [84] , [85] . Thus it seems safe to conclude that dynamic increases in incentive salience are expressed in behavior as well as in neural activation [15] , [86]–[90] .
It is important to acknowledge that each experiment above is only an imperfect test of the model . The VP sensitization and amphetamine experiments hinged on the assumption that our sequential Pavlovian design decoupled the maximal predictive impact of CS1 from the maximal incentive impact of CS2 . If the assumption were false , the conclusion would be weakened . Likewise , the salt CS study failed to cleanly separate ‘wanting’ from hedonic ‘liking’ , because both were increased together during a natural sodium appetite . However , each experiment also carries strengths to counter these weaknesses . The sensitization-amphetamine experiment cleanly dissociated ‘wanting’ from ‘liking’ because the dopamine-based activations enhanced ‘wanting’ without at all enhancing sucrose ‘liking’ . The salt experiment cleanly dissociated incentive ‘wanting’ from cached predictions gained by previous learning , without requiring serial CSs , because the previously learned CS value was negative and was dynamically reversed into positive valence at test by a natural specific appetite . Thus some confidence is gained by noting that our conclusions rest on the entire body of evidence rather than on any single experimental result . Comparing versions ( Eqns 3a and 3b ) , we note that Eqn ( 3a ) could be a standard way to express the generic model ( Eqn 3 ) as a multiplicative gain-control mechanism , which generally applies to all univalent cases of modulation . These include enhancements of ‘wanting’ by amphetamine activation of mesolimbic dopamine systems and by permanent sensitization of mesolimbic systems , as shown here , and would also apply to natural cases such as a palatable food becoming more valuable in hunger . It would similarly apply to univalent downshifts in incentive value from good to less good or neutral ( e . g . , physiological satiety or dopamine suppression ) . Further , Eqn ( 3a ) would also apply to changes in aversiveness from bad to worse ( or vice versa ) for fear , disgust or other negative evaluations that involve a negative version of motivation salience , such as changes in active fear or in psychotic paranoia produced by dopamine-blocking drugs or by psychological mood manipulations [53] , [54] , [91] ( Figure 1 ) . Alternatively , Eqn ( 3b ) expresses the generic model as an additive version , which is intended to account for special cases where incentive value actually reverses valence between positive and negative poles . Those include reversal of incentive salience for intense salt from negatively ‘unwanted’ to positively ‘wanted’ during salt appetite , and would also include flips from ‘wanted’ to ‘unwanted’ such as when a sucrose taste is converted from good to bad by aversion conditioning ( i . e . , pairing as CS with nausea as UCS ) , or flips between desire and dread [53] , [90] . Finally , we stress that our model ( Eqns 3 , 3a , 3b ) is not meant as a finished model of incentive salience , but only is an incremental step towards more adequate computational models . Several important challenges remain . One challenge for a future incentive salience model is to better solve the specificity problem involved in the question of ‘what to want most’ ? That problem includes describing how specific types of κ ( e . g . , sodium appetite , hunger , drug sensitization ) interact with specific CSs and their UCS rewards ( e . g . , salt , food , drugs ) to determine the direction of maximal attribution of incentive salience toward a particular target . A related problem concerns the control of how sharply ‘wanting’ is focused by amygdala-related systems on one CS motivational magnet [48] , [85] , or on one UCS target in a winner-take-all fashion ( as when an addict excessively ‘wants’ only drugs ) , or instead is spread somewhat over several targets ( as when the addict also excessively ‘wants’ to gamble or engage in sex ) [8] . Another challenge is to model the relation of incentive salience ‘reboosting’ ( via incremental pairings of CS and UCS ) to dynamic modulation ( as shown here for a specific CS ) . A final challenge is to better capture the computational differences between Pavlovian-based ‘wanting’ described here versus tree-based cognitive goal systems and cached-based habit learning systems , and to better understand the conditions that determine whether those three systems cohere or diverge . To summarize , we have proposed a computational model of incentive salience as a motivational gating mechanism that dynamically responds to post-learning shifts in physiological states when encoding a relevant CS for reward . Our computation of incentive salience integrates a current change in physiological state with previously learned associations between a CS and its state-relevant UCS reward to generate ‘wanting’ in a dynamic and reversible fashion . The computation of incentive salience outlined here implies that cue-triggered ‘wanting’ amounts to activating associations that exist between CS and UCS , and then dynamically recomputing motivational value based on current physiological state to generate the motivational magnet property of a reward cue [2] , [3] , [7] , [8] . In natural appetites , like salt appetite or food hunger , the dynamic modulation is adaptive , and guides motivated behavior towards an appropriate incentive without need for stable experience-gained knowledge of the goal . In addicts , amplified motivation may maladaptively pull the addict like a magnet towards compulsively ‘wanted’ drugs , and so make it harder to escape from the addiction [8] , [13] , [98] .
|
Reward cues are potent triggers of desires , ranging from normal appetites to compulsive addictions . Food cues may trigger a sudden desire to eat before lunch , and drug cues may trigger even a ‘recovered addict’ to relapse again into drug taking . But learned cues are not constant in their motivating power . Food cues are more potent when you are hungry , and drug cues may become overwhelmingly potent to an addict who tries to take ‘just one’ drink or hit , precipitating an escalating binge of further relapse . These changes in cue-triggered desire produced by a change in biological state present a challenge to many current computational models of motivation . Such modulation can even be unlearned ( though the modulation interacts with cues acquired through learning ) , in that the modulation instantly follows a physiological or neurobiological change ( hunger , drug hit , etc . ) , altering the cue's ability to trigger desire for a relevant reward . Here we demonstrate concrete examples of instant modulation and propose how to build computational models of cue-triggered ‘wanting’ to better capture the dynamic interaction between learning and physiology that controls the incentive salience mechanism of motivation for rewards .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"neuroscience/behavioral",
"neuroscience",
"neuroscience/cognitive",
"neuroscience",
"neuroscience/neural",
"homeostasis",
"neuroscience/psychology",
"neuroscience/theoretical",
"neuroscience"
] |
2009
|
A Neural Computational Model of Incentive Salience
|
Although it is known that mechanical forces are needed for normal bone development , the current understanding of how biophysical stimuli are interpreted by and integrated with genetic regulatory mechanisms is limited . Mechanical forces are thought to be mediated in cells by “mechanosensitive” genes , but it is a challenge to demonstrate that the genetic regulation of the biological system is dependant on particular mechanical forces in vivo . We propose a new means of selecting candidate mechanosensitive genes by comparing in vivo gene expression patterns with patterns of biophysical stimuli , computed using finite element analysis . In this study , finite element analyses of the avian embryonic limb were performed using anatomically realistic rudiment and muscle morphologies , and patterns of biophysical stimuli were compared with the expression patterns of four candidate mechanosensitive genes integral to bone development . The expression patterns of two genes , Collagen X ( ColX ) and Indian hedgehog ( Ihh ) , were shown to colocalise with biophysical stimuli induced by embryonic muscle contractions , identifying them as potentially being involved in the mechanoregulation of bone formation . An altered mechanical environment was induced in the embryonic chick , where a neuromuscular blocking agent was administered in ovo to modify skeletal muscle contractions . Finite element analyses predicted dramatic changes in levels and patterns of biophysical stimuli , and a number of immobilised specimens exhibited differences in ColX and Ihh expression . The results obtained indicate that computationally derived patterns of biophysical stimuli can be used to inform a directed search for genes that may play a mechanoregulatory role in particular in vivo events or processes . Furthermore , the experimental data demonstrate that ColX and Ihh are involved in mechanoregulatory pathways and may be key mediators in translating information from the mechanical environment to the molecular regulation of bone formation in the embryo .
It is widely accepted that there is a relationship between the morphology of skeletal structures and the mechanical forces acting upon them . Such a relationship begins in the embryo where the importance of muscle for normal bone formation has been clearly demonstrated [1] , [2]; however , it is still not understood how biophysical stimuli are interpreted and integrated with the genetic regulatory mechanisms guiding bone development . Presumably gene activity within the skeletal tissues is influenced by mechanical stimulation but there is very limited information on how this might occur in the embryo . Up- and down-regulation of gene expression due to mechanical stimulation has been demonstrated under certain cell culture conditions , and these genes have been called mechanosensitive genes [3] . Most experiments revealing mechanosensitivity have placed cells under mechanical stimulation in culture and subsequently performed analyses to quantitatively compare the expression of many genes between stimulated and control cells , for example using microarray analysis ( e . g . , [4] , [5] ) . Using such an approach , hundreds of potential mechanosensitive genes can be identified simultaneously; however , these experiments do not demonstrate the mechanosensitivity of a gene in an in vivo context . To establish that a gene plays a mechanoregulatory role during a particular process it is necessary to examine the sensitivity of the gene to mechanical stimulation in vivo . It is , however , more challenging to examine candidate genes in an in vivo context . To date , the study of Kavanagh et al . [6] is unique in demonstrating a mechanoregulatory role for a gene during embryonic development in vivo by altering the mechanical environment . These authors examined the expression patterns of three signalling molecules which are implicated in regulating joint formation; growth and differentiation factor 5 ( GDF-5 ) , fibroblast growth factor-2 ( FGF-2 ) and FGF-4 in control and immobilised chick embryonic hindlimbs and showed that joint line FGF-2 expression was diminished in immobilised limbs , while the expression of the other two genes in the joint line was unaffected . They concluded that FGF-2 has a direct mechanoregulatory role in the cavitation process . Another approach has been to use computational modelling to identify candidate mechanosensitive genes , where regions predicted to be under high mechanical stimulation are correlated with the expression of certain genes . Henderson et al . [7] used a 2-D finite element model to predict patterns of growth-related stresses and strains generated during the growth of a skeletal condensation for comparison with in vivo expression patterns of “chondrogenic genes” and “osteogenic genes” . By comparing patterns of biophysical stimuli with gene expression data from transverse sections , they proposed that predicted patterns of pressure correspond with expression patterns of chondrogenic genes and that predicted patterns of strain correspond with patterns of osteogenic genes . Their model focussed exclusively on growth related biophysical stimuli and did not , therefore , examine the effect of embryonic muscle contractions . Considering embryonic bone formation specifically , a number of genes involved in key steps have been identified as mechanosensitive in in vitro cell culture assays [3] , [8] . These include genes encoding Collagen X ( ColX ) , Fibroblast Growth Factor receptor2 ( FGFr2 ) , Indian hedgehog ( Ihh ) and Parathyroid hormone-related protein ( PTHrP ) . ColX encodes a structural protein synthesised by hypertrophic chondrocytes [9] that has been identified as playing a role in matrix mineralization [10] , and was shown to be upregulated in in vitro cultures of bovine chondrocytes under cyclic tension and cyclic hydrostatic pressure [11] and in ex vivo mechanical stimulation of neonatal rabbit distal femoral condyle explants [12] . FGFr2 is a positive regulator of chondrocyte proliferation [13] , and has been shown to be downregulated following in vitro four point bending of MC3T3-E1 preosteoblasts [5] and upregulated in in vitro mechanical stimulation of bone marrow stromal cells [14] . Ihh is also a positive regulator of proliferation [15] , and controls the onset of chondrocyte hypertrophy primarily via PTHrP [16] . Ihh signalling from the proliferative region is necessary to induce the differentiation of the perichondrium into an osteogenic tissue from which the first osteoblasts will differentiate [15] . PTHrP signalling has been shown to negatively regulate the switch from a proliferative immature chondrocyte to a post-proliferative mature hypertrophic chondrocyte [17] . Ihh and PTHrP have been shown to be upregulated by mechanical stimulation; Ihh and PTHrP in in vivo mechanical stimulation of rat mandibular condyles [18] , [19] , Ihh in in vitro cyclic mechanical stimulation of embryonic chick chondrocytes [20] and PTHrP in in vitro cyclic mechanical stimulation of rat growth plate chondrocytes [21] . In this paper , we hypothesise that mechanical forces influence embryonic bone formation by regulating expression of mechanosensitive genes . To test this hypothesis , the involvement of four genes in transducing mechanical information from spontaneous muscle contractions during ossification was assessed; these are ColX , FGFr2 , Ihh and PTHrP . The genes were selected for this study based on their importance for bone formation and evidence of their mechanosensitivity in vitro . Using a novel approach , the potential in vivo mechanosensitivity of these genes is initially assessed using computationally derived data on the biophysical environment . The candidate genes were first examined by correlating their expression patterns with patterns of biophysical stimuli across stages of development when ossification begins . We carried out a detailed analysis of expression of the 4 candidate genes and , by using the results of finite element analyses based on 3-D rudiment morphologies and realistic muscle loading schemes described in a previous paper [22] , we could compare the complex and time-dependant patterns of biophysical stimuli induced by embryonic muscle contractions with gene expression patterns at several timepoints . To corroborate the correlations found , the direct response of both the genes and the patterns of biophysical stimuli to a perturbation in the mechanical environment in vivo were examined . If genes whose expression patterns could be shown to have altered expression patterns in a perturbed mechanical environment , then this would provide strong evidence that genes mediate a genetic regulation of the response to mechanical information during embryonic bone formation .
Morphological and gene expression analyses were carried out on the tibiotarsal rudiment in the hindlimb of the embryonic chick . Dissected embryos were staged according to the Hamburger and Hamilton ( HH ) system [23] . Three stages were chosen for analysis; HH30 , HH32 and HH34 , corresponding to roughly 6 , 7 and 8 days of incubation , spanning the initiation of osteogenesis in the tibiotarsus . The BBSRC ( Biotechnology and Biological Sciences Research Council , U . K . ) ChickEST Database ( http://www . chick . manchester . ac . uk/ , last accessed September 2008 ) and bank of Expressed Sequence Tags ( ESTs ) from the chick genome were used as a source of cDNA clones from which to generate specific RNA expression probes for the genes of interest . The database was searched for ESTs corresponding to each gene and two ESTs were selected for each based on confirmation of perfect alignment with the gene of interest following a Basic Local Alignment Search Tool ( BLAST [24] ) analysis through the National Center for Biotechnology Information ( NCBI , http://www . ncbi . nlm . nih . gov/BLAST/ , last accessed September 2008 ) , and on the length of the EST and its position within the cDNA of the gene of interest . ESTs of 0 . 5–1 . 0 kb were preferred . The probe generated for ColX was produced from chEST 62e2 and aligns with nucleotides 1605–2320 on Genbank sequence ref M13496 . 1 . The probe generated for FGFr2 was produced from chEST 699l24 and aligns with nucleotides 1967–2716 on Genbank ref NM_205319 . The probe generated for PTHrP was produced from chEST 533c1 and aligns with nucleotides 68–734 on Genbank ref AB175678 . The Ihh cDNA clone used for probe production was a gift from C . Tickle ( Dundee ) and corresponds to nucleotides 2–547 on Genbank ref NM_204957 . The probe generated for Scleraxis was produced from chEST 654f15 and aligns with nucleotides 416–1109 on Genbank sequence ref NM_204253 . 1 . Each EST clone was sequenced to verify identity . Plasmid DNA carrying the EST of interest was linearized with appropriate restriction enzymes ( EcoR1 or Not1 ) . Antisense and sense digoxigenin-labelled RNA probes were transcribed in vitro from 1 µg of linearized plasmid using T7 and T3 promoter sites ( according to insert orientation ) in the pBluescript II KS+ vector ( all components for in vitro transcription from Roche , Germany ) . DNA template was degraded by incubation of probes with RNase free DNase ( Roche ) . The probes were then purified on G25 columns ( Amersham Biosciences , USA ) according to the manufacturer's instructions . Probe concentrations were determined by spectophotometry and probes were stored at −20°C . After dissection , limbs selected for in situ hybridisation were fixed in 4% paraformaldehyde ( PFA ) in PBS over night , and dehydrated through a series of methanol/PBT ( PBT = 0 . 1% Triton X-100 in PBS; 25 , 50 , 75%; 1×10 minute ) washes , followed by 2×10 minutes in 100% methanol and stored at −20°C in 30 or 50 ml tubes until needed . On the morning of sectioning , limbs were re-hydrated through a series of methanol/PBT ( 75 , 50 , 25%; 1×10 minute ) washes at 4°C . After 2×10 minutes washes in PBT , excess tissue surrounding the skeletal rudiments was removed in order to give optimal sectioning performance . The specimens were embedded in 4% Low Melting Agarose/PBS ( Invitrogen , UK ) . 80 or 100 µm sections were cut in the longitudinal direction with a vibrating microtome ( VT1000S , Leica ) and stored in PBS in 12-well plates . After 2×10 minute washes in PBT , free-floating sections were treated with proteinase K ( 20 µg/ml in PBT ) for 5 minutes at room temperature . Sections were then washed twice in PBT and fixed for 20 minutes in 0 . 2% glutaraldehyde/4% paraformaldehyde ( PFA ) . Fixation was followed by washes ( 3×5 minutes ) in PBT at room temperature , and a further 30 minute PBT wash at 55°C . The sections were then prehybridised at 55°C overnight in a hybridization solution containing 2% blocking reagent ( Roche ) , 50% formamide , 5× SSC ( Saline-sodium citrate buffer ) , 0 . 5% 3-[ ( 3-Cholamidopropyl-[ ( 3-Cholamidopropyl ) dimethylammonio]-1-propanesulfonate ( CHAPS ) , 500 µg/ml Heparin , 1 µg/ml Yeast RNA , 0 . 1% Tween 20 and 5 mM EDTA ( ethylenediamine tetraacetic acid ) ( all components from Sigma , UK , unless otherwise stated ) . Antisense and sense probes were denatured at 80°C for 3 minutes and sections were then incubated at 55°C over 2–3 nights in hybridization solution containing either antisense or sense probe at minimum concentrations of 2 ng/µl . Post-hybridization washes were carried out at 60°C as follows: 2×10 minutes in 2× SSC; 3×20 minutes in 2× SSC/0 . 1% CHAPS; 3×20 minutes in 0 . 2× SSC/0 . 1% CHAPS . The sections were then washed for 2×10 minutes in TNT ( 100 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 1% Tween 20 ) at room temperature and blocked in blocking buffer ( 0 . 1 M maleic acid , 0 . 15 M NaCl , 3% blocking reagent ( Roche ) ) plus 10% goat serum overnight at 4°C . Sections were incubated overnight in fresh blocking buffer ( plus 10% serum ) containing a 1∶1000 dilution of anti-digoxigenin Fab fragments conjugated with alkaline phosphatase ( Roche ) at 4°C , with rocking . The sections were then washed ( 5×1 hour ) at room temperature in TNT and left rocking in TNT over 2 nights at 4°C . On the day the signal was developed , sections were washed in 3 changes of NMT ( 100 mM Tris-HCl , pH 9 . 5 , 100 mM NaCl , 50 mM MgCl2 ) for 15 minutes each . The chromogenic reaction was carried out in NMT containing 17 . 5 µg/ml 4-nitro blue tetrazolium chloride ( NBT; Roche ) and 6 . 25 µg/ml 5-bromo-4-chloro-3-indolyl-phophate ( BCIP; Roche ) . Sections were developed in the dark at room temperature with rocking for 6–8 hours and then fixed in 4% PFA/PBS for 1 hour before mounting on slides with Aquapolymount ( Polysciences , Inc ) . Two sets of immobilisation experiments were performed at different timepoints; named Set A and Set B . In Set A , 120 eggs were assigned as experimental embryos , and 80 as controls , while in Set B , 100 eggs were assigned as experimental embryos , and 80 as controls . The eggs were incubated for 3 days , after which 4 ml of albumen was removed with a syringe so that the embryo would sink lower in the egg and a window could be cut in the shell without rupturing the chorioallantoic membrane . Administration of the neuromuscular blocking agent Decamethonium Bromide ( DMB ) [25] began at either day 5 ( Set A ) or day 6 ( Set B ) of incubation . Embryos assigned to the experimental group were treated daily with 100 µl of 0 . 5% DMB in sterile HBSS ( Hank's Buffered Saline Solution ) , while control embryos were treated with 100 µl of sterile HBSS . Before administration of the drug or saline solution , movement of the embryo was observed and recorded , and dead embryos were discarded . After treatment , the window was sealed using wide plastic tape and the egg returned to the incubator . The treatment was repeated daily until the embryos were harvested at days 8 , 9 and 11 , corresponding to stages HH30–32 at day 8 , HH32–34 at day 9 and HH35–36 at day 11 . All harvested embryos were stained to reveal cartilage and bone using Alcian Blue ( cartilage ) and Alizarin Red ( bone ) using a modification of the protocol of Hogan et al . [26] , with an Alcian Blue concentration of 0 . l% . After staining , the embryos were photographed , and the total length of the tibiotarsus and the length of the bone collar were measured for each specimen . The numbers of control and experimental specimens at days 8 , 9 and 11 are detailed in Table 1 . These parameters were analysed in the statistical package R ( http://www . r-project . org/ , last accessed September 2008 ) , and standard t-tests were performed in order to determine the effect of immobilisation on the morphology of the rudiments . The right limbs of embryos harvested at day 9 were immediately removed for preparation for sectioning and subsequent in situ hybridisation to analyse the expression of candidate mechanosensitive genes . Sections were compared between control and immobilised groups to determine if the altered mechanical environment had an effect on gene expression . As described in detail in Nowlan et al . [22] a set of finite element analyses of embryonic chick hindlimb skeletal rudiments were created for stages HH30 , HH32 and HH34 . At HH30 and HH32 , the rudiments contain cartilage only , while the periosteal bone collar is present at the mid-diaphysis at HH34 . Anatomically accurate rudiment and muscle morphologies were obtained for each stage using Optical Projection Tomography ( OPT ) [27] , and two animals at each stage were analysed to ensure results were stage dependant rather than animal-specific . In order to characterise the biophysical environment in the absence of skeletal muscle contractions , simulations of the immobilised state were carried out and compared with the previously published patterns . Immobilisation using DMB induces rigid paralysis , where muscles are in continuous tetanus [28] . To model this situation , both ventral and dorsal muscle forces were applied simultaneously , as opposed to the situation in a normal embryo , where ventral muscles are active in flexion and the dorsal muscles in extension , as shown in Figure 1 . The magnitude of the force per unit area value was also adjusted in the paralysis simulations . From the study of Reiser et al . [29] , who reported the tension development in twitch and tetanic responses in normal and immobilised chick embryos , we deduced that the tetanic force response from the muscles in the immobilised chicks would be 75% of the twitch response in normal embryos . We therefore adjusted the magnitude of each of the muscle loads to 75% of the previously applied value .
The expression patterns of ColX , FGFr2 , Ihh and PTHrP illustrated in Figures 2–5 are represented schematically by stage ( Figure 6 ) and compared with patterns of biophysical stimuli at longitudinal sections from the finite element analyses of normal ( control ) limbs as described in Nowlan et al . [22] . The predicted fluid velocity and maximum principal strain mid-flexion , ( Figure 6 ) underwent distinctive changes over the three stages examined , both at the ventral and dorsal surfaces ( illustrated as solid red curves ) , and in a longitudinal section through the middle of the rudiment ( Figure 6 , ‘Normal’ sections ) . At HH30 , stimuli levels were at a high level on the perichondrium at the mid-diaphysis of the rudiment . At HH32 , two concentrations of stimuli were apparent proximal and distal to the mid-diaphysis , again on the surface of the rudiment , and by HH34 , these concentrations moved further apart along the length of the rudiment , proximal and distal to the newly formed bone collar [22] . Two genes showed a correlation with the patterns of biophysical stimuli; ColX and Ihh , as their expression followed patterns of events that reflect the stimuli patterns at the same stages . ColX was found to be expressed in the region of hypertrophic chondrocytes and in the region of the perichondrium where bone would soon form , spreading proximally and distally beyond the hypertrophic zone domain at the core at HH30 and HH32 and ahead of the bone collar at HH34 . Therefore its surface expression demonstrated a correlation with the patterns of biophysical stimuli at each of the three stages examined ( Figure 6 ) . In the earlier stages examined , Ihh was expressed uniformly across the pre-hypertrophic zone , in one mid-diaphyseal region at HH30 , and in two bands at increasing distances proximal and distal to the mid-diaphysis at later stages , the expression bands moving proximally and distally in synchrony with the biophysical stimuli at the surface . Expression of Ihh was therefore at the same longitudinal position in the rudiment as , and adjacent to , the peak levels of biophysical stimuli ( Figure 6 ) .
In this study , we set out to test the hypothesis that mechanical forces influence embryonic bone formation by regulating certain mechanosensitive genes . In a first analysis , the expression patterns of four genes; ColX , FGFr2 , Ihh and PTHrP , were characterised and compared with patterns of biophysical stimuli . ColX and Ihh expression patterns correlated with stage-matched patterns of biophysical stimuli , whereas FGFr2 and PTHrP expression patterns did not . This identified ColX and Ihh as potential mechanosensitive genes regulating ossification in the embryo . ColX and Ihh expression patterns followed the same dynamic sequence of events as the patterns of biophysical stimuli , with one peak of expression at the mid-diaphysis at the youngest stage ( HH30 ) , and two peaks progressively more proximal and distal to the mid-diaphysis at HH32 and HH34 . The ColX expression at the surface ( on the perichondrium ) correlates with the locations of peak biophysical stimuli also at the surface , while Ihh expression in the pre-hypertrophic cartilage is at the same longitudinal position in the rudiment as , and adjacent to , the peak levels of biophysical stimuli . In order to corroborate the hypothesis that ColX and Ihh may act as mechanosensitive genes for bone formation in the chick limb , an immobilisation assay was established , where rigid paralysis was induced with the prevention of skeletal muscle contractions . The morphological analysis of the immobilised embryos clearly demonstrated the effect of an altered mechanical environment on skeletal development , with immobilisation leading to shorter tibiotarsi and decreased bone collar formation . Finite Element Analyses of skeletal elements under rigid paralysis indicated a dramatic alteration in patterns of biophysical stimuli both in terms of stage-dependant patterns of biophysical stimuli and magnitudes of stimuli in comparison with the normal case . Aspects of the expression of ColX and Ihh indeed showed altered expression patterns following immobilisation in a proportion of specimens; ( see Figures 8 and 9 ) , corroborating their role in mechanoregulation pathways during ossification in the chick long bone . The identification of Ihh as mechanosensitive in vivo is of particular interest since this gene has been shown to be a key regulator of bone formation in the mouse , and in particular formation of the bone collar , [15] . The elevation of expression close to the periphery of the hypertrophic zone at later stages , as described in the Results section , was precisely the aspect altered in a number of immobilised specimens with an earlier and more obvious peripheral elevation when mechanical stimulation was reduced ( Figure 9 ) – this indicates a more complex regulation of a gene by mechanical forces than a simple up- or down-regulation on the level of expression . Alterations in Ihh expression would affect the switch from a proliferative to a pre-hypertrophic chondrocyte leading to a shorter rudiment [31] , and shorter rudiments were indeed found in the treated limbs . This indicates that mechanical stimulation may play a role in regulating the position and timing of proliferation of immature chondrocytes through Ihh signalling . As the simulations of the immobilised embryos did not exhibit a specific pattern in the region of the pre-hypertrophic chondrocytes that would explain the change in the Ihh gene expression profile , these results also indicate the involvement of one or more molecules interacting with Ihh in one or more mechanoregulatory pathways . Alteration to the expression of ColX was observed in the regions predicted to have highest concentrations of biophysical stimuli ( Figure 8B and 8C ) , where the expression in the perichondrium did not extend proximal and distal to the hypertrophic zone . The Finite Element simulations of the immobilised limbs indicated that peak stimuli levels at the perichondrium at all three stages were dramatically decreased due to rigid paralysis . It is possible that ColX may promote deposition of osteoid on the perichondrium in response to peak levels of mechanical stimulation , which would explain , at least in part , the reduced bone formation in the altered mechanical environment induced by immobilisation . Alternatively , it is possible that expression in the perichondrium does not extend beyond the hypertrophic region due to an increase in the length of the hypertrophic zone . An elevated rate of hypertrophy would lead to a shorter rudiment , as was indeed found in the immobilised specimens in this experiment . However , the altered expression profile of Ihh does not suggest an increase in the number of pre-hypertrophic chondrocytes . Therefore , it is likely that one or more other mechanoregulatory molecules are involved , and this will be a subject for future work . In this study , there was a certain amount of variability in the effect of the neuromuscular blocking agent , and the change in expression patterns of candidate mechanosensitive genes were not seen in all immobilised ( drug-treated ) specimens . This variability is not unexpected since the alteration to muscle contractions is effected by exposure to a pharmaceutical agent where the response to a set dose can vary across individual specimens . A variable response was also evident when movement in the experimental embryos was quantified; while movement was clearly reduced , it was not completely removed in all specimens . However , detectable changes in gene expression were seen for two different genes in multiple specimens , showing a repeatable effect , and the statistically significant decrease in rudiment length and bone formation serves as confirmation of the immobilisation treatment as a means of altering the mechanical environment . The magnitudes of the muscle loads applied for the embryos subjected to rigid paralysis may be an overestimation , because while we have assumed the same volume of muscle in our simulations , it has been widely reported that muscle mass is reduced in immobilised embryos [32] . However , as the models are likely to overestimate the muscle forces in a completely immobilised animal , this will only strengthen our findings of the dramatic effect on the biophysical environment due to paralysis . Another limitation of this research is that late long bone ossification events are significantly different in mammals and birds [33] , where the long bones of birds are formed primarily via periosteal ossification as opposed to a combination of periosteal and endochondral ossification in mammals . However , birds and mammals have the events preceding ossification in common , such as hypertrophy of the chondrocytes and formation of the periosteal bone collar , and therefore genes identified as being mechanosensitive in vivo in the chick are likely to have a similar role in the mammal . The study presented here has revealed the alteration of gene expression as a result of mechanical stimulation . Even though we have identified the in vivo mechanosensitivity of two genes in the developing limb , we do not know what signalling cascades prompted the change in ColX and Ihh expression patterns . For example , focussing on Ihh in particular , while it has been suggested that Ihh regulates proliferation of chondrocytes through the activation of stretch activated channels by mechanical stimulation [20] , it remains to be discovered what transcription factors and other intracellular molecules form the link between stretch activated channels and upregulation of the gene . As ColX and Ihh have now been demonstrated to be involved in mechanosensitive pathways in vivo at specific developmental timepoints , this opens the possibility of dissecting the upstream mechanisms involved in the response . Many researchers have recognized the importance of the interaction between mechanical and biological factors for bone development . A range of biophysical stimuli parameters have been hypothesised to promote ossification , such as low levels of hydrostatic stress and principal strain [34] , local stress and strain magnitudes [35] or low levels of octahedral shear strain and fluid velocity [36]–[39] . The results presented in this study suggest that biophysical stimuli promote ossification through the action of mechanosensitive genes , but it was not possible to determine a magnitude or level of any particular biophysical stimulus necessary for normal mechanosensitive gene expression . Although dramatic decreases in stimuli magnitudes were found between normal and immobilised simulations within stages , the immobilised stimuli magnitudes at HH34 are still higher than normal values at HH30 and HH32 ( Figure 6 ) . This may suggest that the cellular response of cells to mechanical forces in the embryo is not constant across different stages of development . It was also not possible from this study to conclude the precise nature of the mechanical stimulus , ( such as strain or fluid flow ) , causing mechanotransduction . However , with new insight into the interactions between mechanical forces and mechanosensitive genes , computational simulations which incorporate biological and mechanobiological influences on ossification may now be further developed to include specific mechanosensitive genes . Van Donkelaar and Huiskes [40] have , in fact , already developed such a numerical model , simulating the PTHrP-Ihh control loop and its influence on growth plate development . The results of their simulation suggest that the mechanical stimulation of Ihh is likely to have a greater effect than stimulation of PTHrP , a result that was also suggested in this study , by the correlation of gene expression patterns with biophysical stimuli . Our identification of Ihh as being mechanosensitive in vivo further corroborates the findings of van Donkelaar and Huiskes [40] , and demonstrates that , with the identification of other mechanosensitive genes in vivo , and the subsequent development of more complex and detailed simulations , a deeper understanding of how biophysical stimuli are interpreted and integrated with the genetic regulatory mechanisms guiding bone development can be gained . The work presented here has provided a new insight into mechanoregulation of embryonic long bone ossification . This is the first study where finite element analyses of the embryonic limb using anatomically accurate rudiment and muscle morphologies have enabled comparison of predicted biophysical stimuli patterns with gene expression patterns , and the characterisation of the biophysical environment in the growing rudiment when skeletal muscle contractions are prevented . A means of corroborating candidate mechanosensitive genes was proposed and tested , revealing ColX and Ihh as mechanosensitive in vivo during embryonic bone formation , and also identifying them as potential key mediators in translating information from the mechanical environment to the molecular regulation of bone formation in the embryo .
|
While mechanical forces are known to be critical to adult bone maintenance and repair , the importance of mechanobiology in embryonic bone formation is less widely accepted . The influence of mechanical forces on cells is thought to be mediated by “mechanosensitive genes , ” genes which respond to mechanical stimulation . In this research , we examined the situation in the developing embryo . Using finite element analysis , we simulated the biophysical stimuli in the developing bone resulting from spontaneous muscle contractions , incorporating detailed morphology of the developing chick limb . We compared patterns of stimuli with expression patterns of a number of genes involved in bone formation and demonstrated a clear colocalisation in the case of two genes ( Ihh and ColX ) . We then altered the mechanical environment of the growing chick embryo by blocking muscle contractions and demonstrated changes in the magnitudes and patterns of biophysical stimuli and in the expression patterns of both Ihh and ColX . We have demonstrated the value of combining computational techniques with in vivo gene expression analysis to identify genes that may play a mechanoregulatory role and have identified genes that respond to mechanical stimulation during bone formation in vivo .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"developmental",
"biology",
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology",
"cell",
"biology/cell",
"signaling",
"physiology/pattern",
"formation",
"developmental",
"biology/organogenesis",
"cell",
"biology/gene",
"expression"
] |
2008
|
Identification of Mechanosensitive Genes during Embryonic Bone
Formation
|
Semi-mechanistic pharmacokinetic-pharmacodynamic ( PK-PD ) modeling is increasingly used for antimicrobial drug development and optimization of dosage regimens , but systematic simulation-estimation studies to distinguish between competing PD models are lacking . This study compared the ability of static and dynamic in vitro infection models to distinguish between models with different resistance mechanisms and support accurate and precise parameter estimation . Monte Carlo simulations ( MCS ) were performed for models with one susceptible bacterial population without ( M1 ) or with a resting stage ( M2 ) , a one population model with adaptive resistance ( M5 ) , models with pre-existing susceptible and resistant populations without ( M3 ) or with ( M4 ) inter-conversion , and a model with two pre-existing populations with adaptive resistance ( M6 ) . For each model , 200 datasets of the total bacterial population were simulated over 24h using static antibiotic concentrations ( 256-fold concentration range ) or over 48h under dynamic conditions ( dosing every 12h; elimination half-life: 1h ) . Twelve-hundred random datasets ( each containing 20 curves for static or four curves for dynamic conditions ) were generated by bootstrapping . Each dataset was estimated by all six models via population PD modeling to compare bias and precision . For M1 and M3 , most parameter estimates were unbiased ( <10% ) and had good imprecision ( <30% ) . However , parameters for adaptive resistance and inter-conversion for M2 , M4 , M5 and M6 had poor bias and large imprecision under static and dynamic conditions . For datasets that only contained viable counts of the total population , common statistical criteria and diagnostic plots did not support sound identification of the true resistance mechanism . Therefore , it seems advisable to quantify resistant bacteria and characterize their MICs and resistance mechanisms to support extended simulations and translate from in vitro experiments to animal infection models and ultimately patients .
Antimicrobial therapy greatly benefits from optimized antibiotic dosage regimens that are supported by pharmacokinetic ( PK ) and pharmacodynamic ( PD ) concepts . Traditionally , the minimum inhibitory concentration ( MIC ) has been the predominant measure of bacterial susceptibility to predict antibiotic efficacy and it still is considered as a ‘gold standard’ for determining the bacterial susceptibility and predicting therapeutic success [1] . As most antibiotics have been available for well over a decade , their development relied heavily on MIC based approaches [2] . Despite its popularity , the MIC is subject to several limitations . It is determined at only one time point ( usually between 16 and 24h ) , at a low initial bacterial inoculum ( i . e . usually in the absence of resistant populations ) , and utilizes constant ( i . e . static ) antibiotic concentrations [1] . Therefore , the MIC neither provides information on the time-course of bacterial killing [2–4] nor on emergence of resistance [5 , 6] . There are several types of mechanisms which contribute to bacterial resistance [7] . Heteroresistance defines the scenario where a small number of resistant bacteria ( e . g . 0 . 0001% [equivalent to 10−6] of the initial inoculum ) are present before initiation of antibiotic therapy [8] . Such a small resistant population is typically not found in standard microbroth MIC tests due to the small initial inoculum used for MIC testing . However , serious infections with a high bacterial burden almost certainly harbor such resistant bacteria at initiation of therapy [7] . Secondly , bacterial adaptation decreases the bacterial susceptibility due to the up-regulation of a resistance mechanism ( such as an efflux pump in response to a quinolone antibiotic or the AmpC β-lactamase enzyme in response to β-lactam antibiotics that bind penicillin-binding protein 4 in Pseudomonas [9] ) . Finally , a phenotypic transition between normal replicating bacteria and tolerant bacteria with a greatly reduced growth rate can result in reduced drug sensitivity [10] . To address some of the limitations of the MIC approach , many static and dynamic in vitro and in vivo infection model studies have assessed the ability of empirical PK/PD indices to predict the efficacy of antibiotics . Such data have proven useful to optimize antibiotic monotherapy regimens for patients [2 , 11 , 12] . The large majority of murine infection model studies only assessed bacterial counts at one time point ( usually 24 h ) and did not assess the time-course of bacterial killing in vivo . In vitro infection model experiments [13] use either static antibiotic concentrations [6 , 10 , 14–17] , simulate the dynamic time-course of antibiotic concentrations observed in patients [5 , 18–20] , or utilize both of these approaches [21–23] . These experimental models provide a wealth of time-course data on bacterial growth and killing [24] . While a series of time-course models for bacterial growth , killing and emergence of resistance has been proposed , it is currently unknown which type of dataset from in vitro studies is required to soundly develop such PK/PD time-course models . We suspected datasets which only contain viable counts of the total bacterial population , but do not contain data on resistant population ( s ) , may be insufficient to distinguish between competing models with different resistance mechanisms . We are also not aware of a systematic simulation-estimation study assessing the bias and precision of parameter estimates from in vitro antibacterial models . Therefore , our objective was to compare the ability of static and dynamic in vitro infection models to identify the PK/PD model with the true resistance mechanism used during simulation and to estimate model parameters accurately and precisely . We used Monte Carlo simulations based on six candidate models and estimated these PK/PD models via importance sampling in the S-ADAPT software which is a robust and one of the latest population estimation algorithms .
All six models contained a logistic growth function to limit growth to a maximum total bacterial population size ( Popmax ) . Model 1 ( M1 ) contained one bacterial population and bacterial killing followed an Emax model . This model was originally proposed for antibiotic PD by Zhi et al . [26] and subsequently used by other investigators [3 , 27] . The differential equation for the number of viable , susceptible bacteria ( S ) was: dSdt=[kg⋅ ( 1−SPopmax ) −kmax⋅CKC50S+C]⋅S ( 1 ) Initial condition ( IC ) : S0 = 10Inoc . The kg is the apparent growth rate constant , C the antibiotic concentration , Popmax the maximum concentration of bacteria , kmax the maximal rate constant of bacterial killing , and KC50S the antibiotic concentration yielding 50% of kmax for susceptible bacteria . The mean generation time ( MGT ) was calculated as the inverse of kg . Model 2 was adapted from Nielsen et al . [10] ( Fig 2 ) . The total bacterial population is comprised of two populations ( i . e . two stages ) , one replicating and sensitive stage and one resting and antibiotic tolerant stage . The total bacterial population is assumed to stimulate the transfer of bacteria from the sensitive stage to the tolerant stage . The transfer from the resting stage to the sensitive stage was assumed to be negligible and fixed to zero following the original publication [10] . Bacteria in the resting stage did not grow and were not killed by the antibiotic . As resting bacteria did not revert back to the proliferating stage , they were only subject to a slow first-order natural death process . Therefore , these tolerant bacteria are ( slowly ) dying which causes a biphasic killing profile at high antibiotic concentrations . In this model , bacteria in the tolerant stage can however not repopulate the replicating population as the reversion to the sensitive replicating stage was assumed to be zero . As bacteria were simulated to be in the early logarithmic growth phase , we assumed that only a small fraction ( 10−5 ) of bacteria in the starting inoculum was in the resting , antibiotic-tolerant stage ( T ) . This choice had very limited impact on the model , as this fraction ( 10−5 ) is equivalent to the number of bacteria that convert from the S to the T stage in approximately 1 min for a 106 CFU/mL inoculum . Both populations were assumed to have the same first-order natural death rate constant ( kdeath ) . The differential equations for the sensitive ( S ) and resting ( T ) population of model 2 are: dSdt=[kg⋅ ( 1−S+TPopmax ) −kmax⋅CKC50+C−kdeath−kfor]⋅SIC:S0=10Inoc-R0 ( 2 ) dTdt=kfor⋅S−kdeath⋅TIC:R0=10Inoc-5 ( 3 ) Where kfor= ( kg−kdeath ) ⋅ ( S+TPopmax ) The kdeath was calculated as the inverse of the mean natural death time ( MDT ) . Model 3 ( Fig 2 ) was derived from Jumbe et al . [28] , Gumbo et al . [29] and Campion et al . [18] . This model included a pre-existing susceptible ( S ) and a pre-existing resistant ( R ) population . Both populations did not interconvert . The initial condition of the resistant population is the mutation frequency ( mutf ) multiplied by the total inoculum and the initial condition of the susceptible population is the remainder of bacteria . Both populations were assumed to have the same maximal killing rate constant ( kmax ) and the same growth rate constant ( kg ) . These populations differed however in their drug concentration yielding 50% of kmax . The drug concentration causing 50% of kmax was smaller for the susceptible population ( KC50S ) than the drug concentration causing 50% of kmax for the resistant population ( KC50R ) . This yields the following differential equations for model 3: dSdt=[kg⋅ ( 1−S+RPopmax ) −kmax⋅CKC50S+C]⋅SIC:S0=10Inoc⋅ ( 1-10mutf ) ( 4 ) dRdt=[kg⋅ ( 1−S+RPopmax ) −kmax⋅CKC50R+C]⋅RIC:R0=10Inoc-mutf ( 5 ) In comparison to model 3 , model 4 ( M4 ) contained an additional bi-directional inter-conversion between the susceptible and resistant population ( Fig 2 ) . Model 4 was derived from Jusko et al . [30] and Yano et al . [16] . The initial inoculum of the resistant population was assumed to be in equilibrium ( i . e . steady-state ) with the susceptible population . Therefore , the initial condition of the resistant population was calculated as CFU0 ∙ kfor / krev and the initial condition of the susceptible population was CFU0 ∙ ( 1—kfor / krev ) . Model 4 was described by the following differential equations: dSdt=[kg⋅ ( 1−S+RPopmax ) −kmax⋅CKC50S+C]⋅S-kfor⋅S+krev⋅RIC:S0=10Inoc⋅ ( 1-kfor/krev ) ( 6 ) dRdt=[kg⋅ ( 1−S+RPopmax ) −kmax⋅CKC50R+C]⋅R+kfor⋅S−krev⋅R IC:R0=10Inoc⋅kfor/krev ( 7 ) The kfor and krev are the first-order transfer rate constants from the susceptible to the resistant population and vice versa . Model 5 contained one bacterial population with adaptive resistance . An adaptive resistance model has been proposed previously by Tam et al . [31] . In the present study , we propose a new adaptation function that was based on an indirect response model to reflect the situation that bacteria often need to synthesize a protein ( and other biomolecules ) to ( over- ) express a resistance mechanism . The synthesis and turnover of such molecules can be captured by a turnover model . In the present model , the adaptation compartment affected the KC50S to reflect the up-regulation of a bacterial efflux pump . The differential equations for model 5 were: dSdt=[kg⋅ ( 1−SPopmax ) −kmax⋅CKC50S+C]⋅SIC:S0=10Inoc ( 8 ) d ( Adaptation ) dt= ( Smax⋅CSC50+C−Adaptation ) ⋅koutIC:Adaptation0=0 ( 9 ) KC50S=KC50 , base⋅ ( 1+Adaptation ) ( 10 ) The adaptation variable defines the extent of change of KC50S in response to a bacterial alteration ( such as the expression of an efflux pump; e . g . MexXY-OprM in response to an aminoglycoside ) [32 , 33] . The KC50 , base is the antibiotic concentration causing 50% of kmax in absence of adaptation ( e . g . at time zero ) , Smax the maximum fold-increase of KC50S due to adaptive resistance , SC50 the drug concentration that yields 50% of Smax , and kout the first order turnover rate constant for adaptation . The kout was calculated as the inverse of the mean turnover time ( MTTloss ) . In contrast to an earlier model for adaptive resistance [31] , the time to adaptation for the present turnover model is determined by the mean turnover time of adaptive resistance and is thus independent of the antibiotic concentration . Models 3 and 6 both contained two pre-existing populations with different susceptibility . In contrast to model 3 , model 6 contained the same adaptation function as model 5 which affected both the susceptible ( KC50S ) and the resistant ( KC50R ) population in model 6 . These six PD models can be readily expanded by including different mean generation times and different maximum killing rate constants for the susceptible and resistant population . However , for the purposes of this simulation estimation study , the simpler version of these models was preferred to support parameter estimation .
The simulated viable counts profiles for static time-kill experiments yielded two general shapes of profiles ( Fig 3A ) . The first type showed bacterial killing without regrowth ( M1 and M2 ) with model 2 containing a slower terminal phase representing natural death of bacteria in the resting stage . The second group of profiles yielded initial bacterial killing followed by regrowth due to a resistant bacterial population ( M3 and M4 ) , adaptation ( M5 ) , or both ( M6 ) . Fig 3B shows the viable count profiles simulated under dynamic conditions where regrowth is in part due to low antibiotic concentrations towards the end of the 12-h dosing intervals . Model selection . Each column in Table 3 refers to one true model used for simulation under static or dynamic conditions . The lines in Table 3 show the frequency of selecting the respective model as the best model . If M1 was the true model , both static and dynamic conditions identified M1 as the true model in 93% or 89% of the cases . For the model with one population with a resting stage ( M2 ) , model 2 was correctly identified as the best model in 96% of cases for the static scenario but only in 8% of the cases for the dynamic scenario . When model 3 was used as true model for simulations , static conditions correctly identified M3 as the best model in 82% and dynamic conditions in 97% of the cases ( Table 3 ) . Interestingly , when the model with two populations and a slow inter-conversion ( M4 ) was the true model used for simulations , M3 was incorrectly selected as the best model in 81% ( static ) or 96% ( dynamic setting ) of the cases . Identification of both models with adaptive resistance ( M5 and M6 ) as the true model was only achieved in 10% to 54% of the cases under both scenarios ( Table 3 ) . Bias and imprecision of parameter estimates . Table 4 ( all six models ) and Fig 4 ( models 1 , 3 and 5 ) compare the true parameter values with the median parameter estimates under static and dynamic conditions ( based on the 100 datasets for each model and case ) . Overall , the precision of parameter estimates tended to be better for static compared to dynamic conditions . The median estimates were within 10% of the true value and the imprecision was <20% CV for most parameters of M1 and M2 under both static and dynamic conditions ( Table 4 ) . A noticeable exception was the estimated mean time of natural death ( MDT ) of resting bacteria in M2 which was considerably biased by 326% ( estimated: 1 , 279 min vs . true: 400 min ) in the dynamic setting and biased by 20% ( estimated: 321 min vs . true: 400 min ) in the static setting . For the model with a susceptible and resistant population without inter-conversion ( M3 ) , the vast majority of median estimates were within 10% of the true value with exception of KC50S ( estimated 34% higher ) and KC50R ( estimated 82% higher than the true value ) in the static setting . Both for models 3 and 4 , the dynamic setting provided less biased parameter estimates . However , the slow inter-conversion rate constants ( kfor and krev ) of M4 were difficult to estimate under both settings . For models with adaptive resistance ( M5 and M6 ) , most model parameters were estimated close to their true values and with reasonable precision . However , the parameters related to the adaptation process ( i . e . Smax , SC50 and MTTloss ) were considerably biased and estimated with poor precision for models M5 and M6 under both scenarios . The estimates for Smax and SC50 may be considered reasonable , since the mean turnover time for adaptive resistance was chosen to be 20 h and therefore almost as long as the experimental duration of 24 h for the simulated static time-kill studies . Thus , precise estimation of Smax , SC50 and MTTloss was not expected for the chosen parameter values and experimental design . Impact of biased parameter estimates on viable counts . The viable count profiles predicted from the median estimates under static and dynamic conditions ( Fig 5 ) matched the predicted profiles from the true parameter estimates closely during the first 48 h . For models 3 , 4 and 5 , the deviations were moderate between 48 and 96 h and tended to be larger for the model predictions under static compared to dynamic conditions ( Fig 5 ) . Predictions were slightly better for the two population model without adaptation ( M3 ) than those for the model containing one population with adaptation ( M5 ) . Although some of the parameter estimates were biased for the more complex model M6 ( two populations with adaptation ) , the predictions matched the observations over 96 h closely ( Fig 5 ) .
During the last five decades , a considerable variety of structures for models with irreversible drug effects has been proposed in antimicrobial and antineoplastic chemotherapy [3 , 7 , 42] . These published models include both empiric descriptions of viable count profiles and mechanism-based models . The latter models were developed to characterize relevant aspects of the mechanisms of antibiotic action , bacterial resistance and tolerance for antibiotic mono- and combination therapy and are highly useful to predict the time-course of bacterial growth , killing and resistance and to thereby optimize antibiotic dosage regimens . The vast majority of these antibacterial PK/PD models [3 , 7 , 42] were developed using data on the total bacterial population and did not model viable counts from antibiotic containing agar plates . However , several models co-modelled both the total and resistant populations [28 , 43 , 44] . In this context , it seems surprising that no systematic simulation-estimation study has yet been published to assess the ability to distinguish between competing antimicrobial PD models with different resistance mechanisms . This lack of knowledge affects the vast majority of mathematical models in antibacterial PD . We addressed this gap by performing Monte Carlo simulations with in total 1 , 200 simulated datasets that were estimated using six relevant structural models and two common designs for in vitro infection models . These models reflected genotypically stable resistance mechanisms ( model M3 ) , phenotypic resistance ( i . e . adaptation for M5 and persisters for M2 ) , inter-conversion between bacterial populations ( M4 ) , or multiple of these mechanisms ( M6 , Fig 2 ) . The 1 , 200 datasets were estimated for both the true model and the other five models ( i . e . 6 models x 1 , 200 datasets = 7 , 200 population estimations in total ) to assess the ability to distinguish between competing models . While models M1 and M2 yielded bacterial killing and death without regrowth ( Fig 3A ) , models M3 to M6 could all describe viable count profiles with initial killing followed by regrowth due to emergence of resistance in the presence of constant antibiotic concentrations . It was therefore interesting to assess , whether a robust population PK/PD estimation algorithm ( i . e . importance sampling ) could adequately distinguish between competing models . Despite the use of one of the latest population modeling algorithms , this simulation-estimation study showed that standard statistical criteria could only identify the true model under both static and dynamic conditions in more than 80% of the cases for models M1 and M3 ( Table 3 ) . Importantly , M3 was incorrectly selected as the best model in the large majority of cases even if models M4 , M5 or M6 were the true model used during simulation . For datasets that only contained viable count profiles of the total population , statistical modeling criteria could therefore not reliably identify the true model in case of regrowth due to bacterial resistance . To provide recommendations for the design of future experiments , the MIC can be related to the KC50S as shown previously [24] . The expected KC50S can then be used to guide the concentrations range to be evaluated in in vitro time-kill studies . An a priori choice would be to assess antibiotic concentrations below and above the KC50S ( for instance 2-fold dilutions from 0 . 125 to 32 times the KC50S ) . Determining the MIC at the end of the study ( e . g . 24 h ) experimentally yields valuable information about the extent of resistance development . This information could be subsequently used in additional experiments to assess higher antibiotic concentrations . Additionally , the level of resistance at the end of the experiment will also inform the mechanism ( s ) of resistance and therefore support the choice of the PK/PD model . Quantitative viable count data of the resistant population ( s ) from antibiotic-containing agar plates at 0 and 24 h , for example , can provide experimental evidence to accept or reject several candidate models ( Fig 2 ) . If one observes resistant bacteria on antibiotic-containing agar plates ( containing e . g . 3x the MIC of the antibiotic ) at 0 h , one can reject models M1 , M2 , and M5 , as those models assume the absence of resistant bacteria at time zero . Mutation frequency studies at a high bacterial density are expected to support calculation of the likelihood of pre-existing resistant mutants [7] . Quantifying and modeling resistant bacteria over time would further enhance the ability to distinguish between competing models [28 , 43 , 44] . An in-depth analysis of datasets containing one or multiple observations for resistant bacteria is beyond the scope of the present work . While this is a potential limitation of this simulation-estimation study , experimental data on the presence or absence of pre-existing resistant bacteria facilitated PK/PD model selection in previous studies [28 , 43 , 44] . The difficulty to select the most appropriate mechanism of resistance based on modelling methods alone is also supported by experience from our previous study on P . aeruginosa exposed to static concentrations of ciprofloxacin [6] . In this study , we leveraged insights on the presumably most relevant mechanism of resistance to select the final model for ciprofloxacin . Despite considerable bias for some parameter estimates , the discrepancies between predicted and actual viable count profiles ( Fig 5 ) were limited and may possibly be considered acceptable . This applies particularly for the small discrepancies during the first 24 h to 48 h which is likely the most critical time in the management of infections in critically-ill patients . Model predictions over longer time periods ( i . e . extrapolation ) led to more biased predictions , as expected . Our predictions were based on the median of the parameters values . As these simulations did not account for parameter imprecision , the discrepancies between the predicted and the actual viable count profiles are likely larger for some sets of estimated parameters . Bias tended to be less for some parameters under the dynamic compared to the static experimental setting for models with heteroresistance or adaptation ( M3 , M4 , M5 and M6; Table 4 and Fig 4 ) . As expected , the dynamic setting yielded less precise parameter estimates most likely due to the considerably smaller number of observations for the dynamic setting ( containing 4 curves per dataset ) compared to the static setting ( containing 20 curves per dataset ) . In practice the statistical gain of the dynamic design can also be offset by the significantly increased workload for dynamic experimental conditions . Static concentration time-kill studies [13] are efficient and cost-effective and allow studying a large range of antibiotic concentrations . While 24 h static time-kill studies represent the most common experimental duration , longer experimental durations could have allowed us to increase the likelihood of mathematically identifying the model with the true resistance mechanisms and to better predict antimicrobial efficacy over longer periods of time . Most published studies did not exchange the broth medium regularly ( e . g . every 24 h ) and therefore toxic bacterial metabolites may accumulate and nutrients get depleted over time . Also , degradation ( e . g . of β-lactam antibiotics ) over longer experimental conditions would need to be accounted for experimentally as we published previously [33] . Overall , performing static concentration time-kill studies over more than 24 h is fully feasible , but requires an increased amount of work . Dynamic in vitro infection models such as the one-compartment and hollow-fiber system can mimic human PK [4] , by changing drug concentrations and turnover of fresh broth medium using various pumps . The control of these flow rates permits to simulate different half-lives for one or multiple drugs and also provides washout of toxic bacterial metabolites . Therefore , these dynamic experiments are often run over multiple days to week and longer [31 , 45] and typically use multiple dosing [46] . These dynamic in vitro models require a significantly enhanced workload and therefore complement and extend static concentration time-kill studies for translation to animal studies and ultimately to patients . Some limitations of our study came from the necessity to select PD parameters characterizing the simulated pathogen , to choose an experimental design for static and dynamic kill-curves and to set the initial values for parameter estimations . Our chosen parameter values adequately characterized the concentration-effect relationship for each studied PD model and were selected based on biological plausibility and our experience from experimental datasets . Despite a clear concentration-effect relationship in the simulated viable count profiles for each dataset , it was difficult to impossible to mathematically identify the true resistance mechanism based on viable counts of the total population . It is possible that other sets of parameter estimates used during simulation would have yielded a higher likelihood to identify the model with the true resistance mechanism . This presents a potential limitation of the present study . Moreover , the present study contained simulation-estimation scenarios for static and dynamic in vitro infection models . Our simulated dynamic in vitro model only assessed a scenario with one half-life , one dose level , and one dosing interval . Dynamic infection models with multiple doses , different dosing intervals , and potentially a range of relevant half-lives would provide a more informative dataset which may have supported the identification of the PD model with the true resistance mechanism . While it is a potential limitation of our study that we did not evaluate more dynamic infection model studies , most current papers on antimicrobial PD models use static time-kill experiments to define the concentration-effect relationship and select the PD model and its resistance mechanism . Finally , we did not assess the sensitivity of the final parameter estimates and the model selection towards the choice of the initial estimates . In our previous work , we found the importance sampling algorithm in S-ADAPT to be robust and efficient despite the use of poor ( i . e . 10-fold too high or 10-fold too low ) initial estimates for every structural model parameter in a complex PK/PD model [7] . In summary , for datasets based only on the total bacterial population , standard statistical modeling criteria failed to correctly identify the PD model with the true resistance mechanism ( s ) in the large majority of cases . These datasets did not contain data on antibiotic-resistant bacterial populations . This finding is highly important , as most published models in antibacterial PD were developed based only on data on the total bacterial population . For our simulation scenarios , dynamic infection models tended to provide more accurate parameter estimates than static concentration time-kill studies for some parameters . Static time-kill studies yielded more precise parameter estimates compared to dynamic models likely due to the larger number of profiles per datasets . For both static and dynamic conditions , parameters related to adaptive resistance and interconversion of bacterial populations tended to be poorly estimated . Interestingly , predicted viable count profiles over the experimentally studied duration ( i . e . 24 to 48 h ) were reasonably accurate despite biased parameter estimates . Simulations over longer durations ( i . e . extrapolations ) tended to show more pronounced mispredictions and should be interpreted conservatively . Overall , it seems highly beneficial to utilize quantitative viable count data of resistant populations and characterize their MICs and resistance mechanisms to support the choice of the most appropriate PD model for bacterial resistance .
|
Mathematical models are increasingly used for analysis and interpretation of in vitro efficacy results of antimicrobial drugs . Various models are employed in the scientific literature and it seems that they are equally able to describe the observed data . The aim of the present study was to compare different models in various experimental designs and with different resistance mechanisms of bacteria . For that purpose we have generated experimental data through Monte-Carlo simulations and then used six different mathematical models to analyze these results . We showed that statistical comparison of models did not allow determining which was the true mechanism of resistance , i . e . the one used for the simulation step . Moreover mathematical parameters for bacterial resistance were estimated with bias and with a low precision except for the simpler cases . This suggests that the choice of the mathematical model for data analysis should be guided by experimental characterization of the bacterial mechanism of resistance .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"drugs",
"experimental",
"design",
"microbiology",
"mathematical",
"models",
"simulation",
"and",
"modeling",
"research",
"design",
"antibiotic",
"resistance",
"mathematics",
"statistics",
"(mathematics)",
"antibiotics",
"pharmacology",
"research",
"and",
"analysis",
"methods",
"antimicrobial",
"resistance",
"mathematical",
"and",
"statistical",
"techniques",
"monte",
"carlo",
"method",
"differential",
"equations",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"statistical",
"methods"
] |
2016
|
Distinguishing Antimicrobial Models with Different Resistance Mechanisms via Population Pharmacodynamic Modeling
|
We have developed a transcriptome-wide approach to identify genes affected by promoter CpG island DNA hypermethylation and transcriptional silencing in colorectal cancer . By screening cell lines and validating tumor-specific hypermethylation in a panel of primary human colorectal cancer samples , we estimate that nearly 5% or more of all known genes may be promoter methylated in an individual tumor . When directly compared to gene mutations , we find larger numbers of genes hypermethylated in individual tumors , and a higher frequency of hypermethylation within individual genes harboring either genetic or epigenetic changes . Thus , to enumerate the full spectrum of alterations in the human cancer genome , and to facilitate the most efficacious grouping of tumors to identify cancer biomarkers and tailor therapeutic approaches , both genetic and epigenetic screens should be undertaken .
It is now well established that loss of proper gene function in human cancer can occur through both genetic and epigenetic mechanisms [1 , 2] . The number of genes mutated in human tumor samples is being clarified . Recently , Sjöblom et al . [3] sequenced 13 , 023 genes in colorectal cancer ( CRC ) and breast cancer , and estimated an average of 14 significant mutations per tumor , suggesting that a relatively small number of genetic events may be sufficient to drive tumorigenesis . In contrast , the full spectrum of epigenetic alterations is not well delineated . The best-defined epigenetic alteration of cancer genes involves DNA hypermethylation of clustered CpG dinucleotides , or CpG islands , in promoter regions associated with the transcriptional inactivation of the affected genes [2] . These promoters are located proximal to nearly half of all genes [4] and are thought to remain primarily methylation free in normal somatic tissues . The exact number of such epigenetic lesions in any given tumor is not precisely known , although a growing number of screening approaches , none covering the whole genome efficiently , are identifying an increasing number of candidate genes [5–13] . Given the large number of potential target promoters present in the genome , we hypothesized that many more hypermethylated genes await discovery . Herein , we describe a whole human transcriptome microarray screen to identify genes silenced by promoter hypermethylation in human CRC . The approach readily identifies candidate cancer genes in single tumors with a high efficiency of validation . By comparing the list of candidate hypermethylated genes with mutated genes recently identified in CRC [3] , we establish key relationships between the altered tumor genome and the gene hypermethylome . Our studies provide a platform to understand how epigenetic and genetic alterations drive human tumorigenesis .
Our first step towards a global identification of hypermethylation-dependent gene expression changes was made by comparing , in a genome-wide expression array-based approach , wild-type HCT116 CRC cells with isogenic partner cells carrying individual and combinatorial genetic deletions of two major human DNA methyltransferases ( Figure 1A ) [14] . Importantly , in the DNMT1 ( −/− ) DNMT3B ( −/− ) double knockout ( DKO ) HCT116 cells , which have virtually complete loss of global 5-methylcytosine , all previously individually examined hypermethylated genes lacking basal expression in wild-type cells undergo promoter demethylation with concomitant gene re-expression [10 , 14–16] . By stratifying genes according to altered signal intensity on a 44K Agilent Technologies array platform , we observe a unique spike of gene expression increases in the DKO cells when compared to the isogenic wildtype parental cells , or isogenic cell lines in which DNMT1 or DNMT3B have been individually deleted and which harbor minimal changes in DNA methylation ( Figure 1B ) . This minimal change in the DNMT1 ( −/− ) cells may , in part , be due to recently identified alternative transcripts arising from the DNMT1 locus [17 , 18] . We tested our approach using a pharmacologic strategy based on our previous approach [10] , but now markedly modified to provide whole-transcriptome coverage , to identify silenced hypermethylated genes in any cancer cell line . For densely hypermethylated and transcriptionally inactive genes , the DNA demethylating agent 5-aza-2′-deoxycytidine ( DAC ) has a well established capacity to induce gene re-expression [19 , 20] . On the other hand , for these same genes , the class I and II histone deacetylase inhibitor , trichostatin A ( TSA ) will not alone induce reexpression [10 , 21] . We now use this lack of TSA response for such genes to provide a new informatics filter to identify the majority of DNA hypermethylated genes in cancer . After treatment of HCT116 cells with either DAC or TSA ( Figure 1C ) , we identified a zone in which gene expression did not increase with TSA ( <1 . 4-fold ) and displayed no detectable expression in mock-treated cells . Within this zone , we observed a characteristic spike of DAC-induced gene expression that virtually completely encompasses the genes with increased expression in DKO cells ( compare yellow spots in Figure 1D with blue spots in Figure 1B ) . This gene spike is absolutely dependent upon analysis of only genes that fail to respond to histone deacetylase inhibition , underscored by a cluster analysis that shows the close relationship between genes in DKO- and DAC-treated cells with a separate grouping of gene-expression changes after TSA treatment alone or in single knockouts ( Figure 1E ) . These data confirm previous studies covering much less of the genome , and using only treatment of cells with DAC and TSA together , in which genes with dense CpG islands that were reexpressed by TSA harbored only partial or no detectable hypermethylation [10 , 21] . Importantly , a similar spike of gene expression increases could be seen in five additional human CRC cell lines , SW480 , CaCO2 , RKO , HT29 , and COLO320 ( Figure 2A ) , as well as cell lines derived from lung , breast , ovary , kidney , and brain ( unpublished data ) , confirming that this approach works universally in cancer cell lines and identifies overlapping gene sets ( Figure 2C ) . However , it is important to note that—possibly because DAC incorporates into the DNA of dividing cells , and our treatments were performed for only 96 h—sensitivity for detecting the gene increases in the pharmacological approach is reduced in HCT116 cells compared to that seen in DKO cells ( Figure 1D ) . To address the sensitivity with which our new array approach identifies CpG island hypermethylated genes , we first examined 11 genes known to be hypermethylated , completely silenced and reexpressed after DAC treatment in HCT116 cells ( Figure 3A ) . All tested genes remained within the TSA nonresponsive zone ( Figure 3B ) , and the direction of expression changes correlated well in DAC treated and DKO cells ( Figure 3C ) . Importantly , for the DAC increase , five of the guide genes ( 45% ) increased 2-fold or more and three more genes , or a total of 73% , increased 1 . 3-fold or more ( Figure 3D ) . We estimate , then , that we can detect over 70% of DNA hypermethylated genes in a given cancer cell line and we test this hypothesis in studies directly below . Based on the sensitivity differences observed between DKO- and DAC -induced gene increases ( compare Figure 1B and D; also Figure 3B and 3C ) and behavior of the guide genes in the array platform , we designated , within the TSA-negative zone , a top tier ( 2-fold increase or above ) and a next tier of genes ( increasing between 1 . 4- and 2-fold ) to identify hypermethylated cancer genes ( Figure 2B ) . Importantly , we introduced an additional filter for selecting genes from these zones based on their having no basal expression in untreated cells , since this full lack of transcription is characteristic of promoter CpG island methylated genes in cell culture . Indeed , based on these selection criteria , in HCT116 cells , 32 of 35 ( 91% , Figure 4 ) of randomly chosen CpG island–containing genes spanning the top-tier response zone of 532 genes ( Figure 5 ) , and 31 of 48 such SW480 cell genes ( 65% , Figure 6 ) from among 318 top tier genes proved to be CpG hypermethylated as measured by methylation-specific PCR ( MSP ) [22] , and silenced in the cell line of origin as measured by reverse transcriptase PCR ( RT-PCR ) . We also examined the efficiency of discovery for hypermethylated genes in the next tier of DAC-treated HCT116 cells . Of the 1 , 190 genes identified in this region , 17 of 35 ( 49% ) randomly selected genes containing a CpG island were hypermethylated with concordant gene silencing ( Figure 7 ) . Our verification rates then demonstrate around 65% efficiency of our approach , which is close to our original estimate and which is excellent compared to previous screens for identifying new cancer hypermethylated genes [6 , 23] . With this level of verified hypermethylation , we calculate that the hypermethylome in HCT116 cells consists of an estimated 1 , 067 genes and an estimated 579 genes for the SW480 cells ( See Table S1 for a detailed description of calculations ) . The hypermethylome would be estimated to range from 532 genes in CaCO2 to 1 , 389 genes in RKO cells ( Table S1 ) . We next asked whether our top and next-tier regions truly enriched for hypermethylated genes by examining a randomly selected subset of 22 control genes located outside these zones . These genes were located in the responsive TSA zones ( Zones 1 and 2 , Figure S1A ) or below the threshold of DAC responsiveness in the TSA nonresponsive zone ( Zone 3 , Figure S1A ) in HCT116 cells . Of the tested genes , only 9% ( 2 of 22 , Figure S1B ) showed detectable methylation with concomitant gene silencing , confirming the specificity of our approach and validating the criteria we used to establish the top and next-tier approach . We can then predict that for cancer cell lines , with use of our filters , ∼90% of promoter CpG island DNA methylated genes lie in the negative TSA-responsive zone . A fundamental question in cell culture–based approaches is whether they identify genes that are targets for inactivation in primary tumors . To address this , 20 CpG island containing genes from the verified gene lists were randomly selected from the HCT116 top tier ( 17 genes ) , HCT116 next tier ( two genes ) , or SW480 top tier ( one gene ) and analyzed for methylation in a panel of CRC cell lines . All of the tested genes were hypermethylated in two or more cell lines ( Figure 8 ) . We then examined the status of these 20 genes in a panel of 20 to 61 primary colon cancers and 20 to 40 normal-appearing colon tissue samples obtained from cancer-free individuals . Most of the genes ( 65% ) were completely unmethylated or rarely methylated in the normal colonic tissue samples , but were methylated in a vast majority ( 86% ) of the primary tumors ( Figure 8 ) . Of the 20 genes analyzed , 13 genes ( 65% ) satisfied criteria for “tumor-specific methylation” with high-frequency methylation in cell lines , low ( <5% ) or undetectable methylation in normal colon , and frequent methylation in primary tumor samples ( Figure 8 ) . The efficiency of our strategy suggests a discovery rate of approximately one in two for identification of hypermethylated genes in cell lines and approximately one in three for identification of cancer-specific hypermethylated genes . Our estimate of approximately 400 hypermethylated genes per primary tumor now can be matched with predictions of Costello et al . [5] for hypermethylation of CpG islands , based on screening with Restriction Landmark Genomic Scanning approaches . We next tested some parameters for biological significance of two of the genes harboring tumor-specific methylation for their likely importance in primary colon cancers . One , the neuralized homolog ( Drosophila ) ( NEURL ) gene , is located in a chromosome region with high deletion frequency in brain tumors [24] , and its product has been identified as a ubiquitin ligase required for Notch ligand turnover [25–27] . Activation of this key developmental pathway influences cell-fate determination in flies and vertebrates [28 , 29] and activation of Notch , through unknown mechanisms , is thought to play an inhibitory role in normal differentiation during colorectal cancer [30] . The second gene , FOXL2 , belongs to the forkhead domain–containing family of transcription factors implicated in diverse processes including establishing and maintaining differentiation programs [31] . Intriguingly , this gene is essential for proper ovarian development [32] and germline mutations in humans lead to a plethora of craniofacial anomalies and premature ovarian failure [33] . We find both of these genes to be frequently DNA hypermethylated in a panel of colorectal cell lines ( five of nine cell lines for NEURL and seven of nine for FOXL2 , Figure 9A and 9C ) , and bisulfite sequencing revealed methylation of all CpG residues in the central CpG island regions of both genes in HCT116 and RKO cell lines , with complete demethylation in DKO cells ( Figure 9B and 9D ) . For both genes , this hypermethylation perfectly correlated with loss of basal expression and ability to reexpress the genes with DAC treatment ( Figure 9A and 9C ) . Importantly , promoter methylation of both genes , as assessed by bisulfite sequencing ( Figure 9B and 9D ) is absent in normal human colon or rectum , but frequent in primary colon cancers ( Figure 9E and 9F ) , suggesting that hypermethylation arose as a cancer-specific phenomenon , although slight methylation was observed at the FOXL2 locus in normal tissue from aged patients ( unpublished data ) . Finally , the pattern for hypermethylation of the FOXL2 and NEURL genes in cell culture fit with a biology important to a subset of colon cancers . As many as one in eight colorectal cancers , predominantly those from the right side of the colon , harbor a defect in mismatch-repair capacity [34 , 35] , primarily due , in nonfamilial cancers , to inactivation of MLH1 by epigenetic mechanisms [36] . Such tumors belong to a group with high frequency of hypermethylated gene promoters [37 , 38] . The hypermethylation of FOXL2 and , especially , NEURL , aggregate with these tumor types not only among the colon cancer cell lines ( HCT116 , DLD1 , LoVo , RKO , and SW48 ) , but also when analyzed in a series of primary human colon cancers ( Fisher's exact test value of 0 . 024 for FOXL2 and 0 . 001 for NEURL , Figure 9G ) . Initial in vitro studies suggest that both FOXL2 and NEURL might possess tumor-suppressor activity . When overexpressed in colon cancer cell lines , full-length FOXL2 and NEURL ( Figure 10A and 10C ) , generate a 10-fold and 20-fold reduction , respectively , in colony growth of HCT116 cells ( Figure 10C ) , with surviving clones having severely depleted size ( Figure 10B ) , comparable to results obtained with the bona fide tumor suppressor p53 ( Figure 10F ) . Similar results were seen in RKO and DLD1 cells ( Figure 10D and 10E ) , both of which have complete gene silencing at the FOXL2 and NEURL loci . While the precise molecular mechanisms for the growth suppression remains to be determined , Notch signaling has recently been shown to play an important role in differentiation of intestinal crypt cells where deletion of the Notch effector molecule RBPJκ or treatment with a highly selective γ-secretase inhibitor was found to be sufficient for conversion of crypt cells to goblet cells [28 , 29] . Similarly , the closely related FOXL2 transcription factor family member FOXL1 has recently been shown to play a role in epithelial–mesenchymal transition of the intestinal epithelium [39] . While it is clear that genetic and epigenetic mechanisms are both important to initiation and progression of human tumorigenesis , the relative contributions of each of these alterations need to be clarified on a global basis . Studies of classic tumor suppressor genes such as VHL in renal cancer and MLH1 in colon cancer indicate that important cancer genes can have an incidence of inactivation by either genetic or epigenetic mechanisms [36 , 40] . However , a genome-wide analysis to query such relationships has not been performed . In a recent genome-wide sequencing of cancer genes , Sjöblom et al . [3] observed that newly discovered gene mutations in colon and breast cancers generally had a low incidence of occurrence , with 90% of the genes identified harboring a mutation frequency of less than 10% . Furthermore , a typical patient's colon or breast tumor was estimated to have an average of only 14 mutations and there appeared to be little overlap between individual tumors for the newly discovered mutations [3] . These low frequencies raise the question whether alternative mechanisms might account for inactivation of these genes in additional tumors . Obviously , the much higher number of candidate hypermethylated genes we now identify in individual tumors suggests that this epigenetic change might provide an alternative inactivating route to mutations for many tumor suppressor genes . We now show that screening tumors for both genetic and epigenetic changes indicates that this is the case . We first located the 189 newly identified , mutated cancer ( CAN ) genes , described by Sjöblom et al . [3] , within the top and next tiers of our colorectal cancer cell line hypermethylome and found 56 genes present in these zones in one or more of the cell lines . Of these , 45 contained CpG islands . Twenty-six of these 45 genes ( 58% ) , similar to the verification rate for all candidate genes identified as discussed above , proved to be hypermethylated in at least one of the six cell lines , and were selected for further study . Importantly , exactly half ( 13 of 26 genes ) of these genes were expressed at high levels ( Figure 11A ) and were not methylated in normal colon ( Figure 11B ) but were methylated in primary CRC tumors ( Figure 11C ) , giving a frequency of 50% for identification of tumor-specific methylation when starting with genes harboring cell line methylation . We also randomly selected , for verification of methylation and expression status in cell lines , CAN genes that fell primarily in zone 3 of the microarray , that is , within the TSA-negative zone but below the 1 . 4-fold cutoff for stimulation by DAC . As seen earlier for other randomly selected genes in this region , these randomly selected CAN genes had a significantly reduced ( four of 15 , or 27% ) frequency of methylation as compared to the 56 top and next-tier CAN genes discussed above ( Figure S1C ) . Interestingly , however , this rate is much more similar to that for the well-characterized hypermethylated guide genes ( ∼30% as shown in Figure 3A–3C ) than for the other randomly selected zone 3 genes ( 9% , compare Figure S1B and S1C ) , perhaps indicating the importance of epigenetic inactivation of these mutated genes . Indeed , relevant to this point , for the majority of the examined CAN genes within the hypermethylome region , the incidence of hypermethylation is strikingly higher than that for mutations ( Figure 11D ) . Thus , unlike for the mutations in the individual genes , which are restricted to only tumors from a few patients , hypermethylation for the majority of the genes is a shared property between many tumors . These findings of both mutations , and alternatively epigenetic silencing , in these previously uncharacterized genes solidifies their probable roles as tumor suppressor genes .
We describe a gene-expression approach with the capacity to define , for any human cancer type for which representative cell-culture lines are available , a substantial fraction of the cancer gene promoter CpG island DNA hypermethylome . Studies of these genes will contribute to understanding the molecular pathways driving tumorigenesis; provide useful new DNA hypermethylation biomarkers to monitor cancer risk assessment , early diagnosis , and prognosis; and permit better monitoring of gene reexpression during cancer prevention and/or therapeutic strategies [41] . Through use of our approach to analyze mutated genes identified by a genome-wide sequencing strategy , we document that many more epigenetically altered genes than genetically altered genes exist in any given tumor . The importance of this fact emerges in our finding that for newly discovered genes that are affected by both mechanisms , the incidence for hypermethylation of any given gene among colon cancers appears to be generally much higher than for mutations . Interestingly , many of the new genes found by Sjöblom et al . [3] harbored heterozygous mutations and it would be , thus , difficult to predict whether the genes were affected by activating or inactivating events from such data alone . As first suggested by Zardo et al . [42] , our data may clarify , in initial screening studies , the latter category , as promoter DNA hypermethylation and gene silencing often affect genes independent of loss of heterozygosity frequency . Thus , discovery of genes targeted by hypermethylation as an inactivating event should help guide prioritization of genes to study in cancer gene resequencing efforts . Finally , our data indicate that , in any given cancer type , one may markedly underestimate both the full range of gene alterations and associated abnormalities of cellular pathways by failing to screen for both genetic and epigenetic abnormalities . Our findings indicate that assessing both mechanisms for loss of gene function indicates far more sharing among individual colon tumors for pathway disruption than genetic analyses alone would predict . Optimal approaches to grouping of tumors according to molecular alterations in key pathways should , then , depend on defining both genetic and epigenetic gene changes . Thus , our findings should encourage any genome-wide cancer gene screening strategies to include finding DNA hypermethylated genes and prioritizing these to be sequenced for mutations as well as prioritizing newly discovered mutated genes to be studied for promoter methylation .
HCT116 cells and isogenic genetic knockout derivatives were maintained as previously described [14] . For drug treatments , log-phase CRC cells were cultured in McCoy's 5A media ( Invitrogen , http://www . invitrogen . com/ ) containing 10% BCS and 1× penicillin/streptomycin with 5 μM DAC ( stock solution: 1 mM in PBS; Sigma , http://www . sigmaaldrich . com/ ) for 96 h , replacing media and DAC every 24 h . Cell treatment with 300 nM TSA ( stock solution: 1 . 5mM dissolved in ethanol , Sigma ) was performed for 18 h . Control cells underwent mock treatment in parallel with addition of equal volume of PBS or ethanol without drugs . Total RNA was harvested from logphase cells using TRIzol ( Invitrogen ) and the RNeasy kit ( Qiagen , http://www1 . qiagen . com/ ) according to the manufacturer's instructions , including a DNase digestion step . RNA was quantified using the NanoDrop ND-100 ( http://www . nanodrop . com/ ) followed by quality assessment with 2100 Bioanalyzer ( Agilent Technologies , http://www . agilent . com/ ) . RNA concentrations for individual samples were greater than 200ng/μl , with 28S/18S ratios greater than 2 . 2 and RNA integrity of 10 ( 10 scored as the highest ) . Sample amplification and labeling procedures were carried out using the Low RNA Input Fluorescent Linear Amplification Kit ( Agilent Technologies ) according to the manufacturer's instructions . The labeled cRNA was purified using the RNeasy mini kit ( Qiagen ) and quantified . RNA spike-in controls ( Agilent Technologies ) were added to RNA samples before amplification . Samples ( 0 . 75 μg ) labeled with Cy3 or Cy5 were mixed with control targets ( Agilent Technologies ) , assembled on Oligo Microarray , hybridized , and processed according to the Agilent microarray protocol . Scanning was performed with the Agilent G2565BA microarray scanner using settings recommended by Agilent Technologies . All arrays were subject to quality checks recommended by the manufacturer . Images were visually inspected for artifacts , and distributions of signal and background intensity of both red and green channels were examined to identify anomalous arrays . No irregularities were observed , and all arrays were retained and used . All calculations were performed using the R statistical computing platform [43] and packages from Bioconductor bioinformatics software project [44–46] . The log ratio of red signal to green signal was calculated after background subtraction and LoEss normalization as implemented in the limma package from Bioconductor [46] . Individual arrays were scaled to have the same inter-quartile range ( 75th percentile–25th percentile ) . Log-fold changes were averaged over dye-swap replicate microarrays to produce a single set of expression values for each condition . We have deposited primary array data in the GEO database at the National Center for Biotechnology Information ( NCBI ) . RNA was isolated with TRIzol Reagent ( Invitrogen ) according to the manufacturer's instructions . For RT-PCR , 1 μg of total RNA was reverse transcribed using Ready-To-Go You-Prime First-Strand Beads ( Amersham Biosciences , http://www . amersham . com/ ) with addition of random hexamers ( 0 . 2 μg per reaction ) . For RT-primer design we used Primer3 ( http://frodo . wi . mit . edu/cgi-bin/primer3/primer3_www . cgi ) . For MSP analysis , DNA was extracted following a standard phenol-chloroform extraction method . Bisulfite modification of genomic DNA was carried out using the EZ DNA methylation Kit ( Zymo Research , http://www . zymoresearch . com/ ) . Primer sequences specific to the unmethylated and methylated promotor sequences were designed using MSPPrimer ( http://www . mspprimer . org ) . MSP was performed as previously described [22] . All PCR products ( 15 μl of 50-μl total volume for RT-PCR and 7 . 5 μl of 25-μl total volume for MSP ) were loaded directly onto 2% agarose gels containing GelStar Nucleic Acid Gel Stain ( Cambrex , http://www . cambrex . com/ ) and visualized under ultraviolet illumination . Primer sequences and conditions for MSP , bisulfite sequencing , and RT-PCR are available upon request from the authors . Formalin-fixed , paraffin-embedded tissues from primary CRCs were obtained from the archive of the Department of Pathology of the University Hospital Maastricht , Maastricht , The Netherlands and Johns Hopkins University Hospital . Approval was obtained by the Medical Ethical Committees of the University of Maastricht and the University Hospital Maastricht and Johns Hopkins University Hospital . DNA was isolated using the Puregene DNA isolation kit ( Gentra Systems , http://www1 . qiagen . com/ ) . FOXL2 and NEURL methylation was analyzed by nested MSP . MSI analysis was performed by analysis of the BAT-26 mononucleotide repeat . The primer sequences and PCR conditions for the BAT-26 mononucleotide repeat were used as described previously [47] . One million HCT116 , RKO , or DLD1 cells were plated in six-well dishes ( Falcon ) and transfected with 5 μg of plasmid ( pIRES-Neo3 , Invitrogen ) using Lipofectamine 2000 according to the manufacturer's instructions . Following a 24-h recovery period , selection in 4 μg/ml gentamycin- ( Invitrogen ) containing complete medium was performed for 10 d . Staining , visualization , and counting of triplicate wells were performed as previously described [48] .
We have deposited primary array data in the GEO database at the NCBI ( http://www . ncbi . nlm . nih . gov/geo/ ) . The accession numbers for the array experiments described in the paper are: GSM107602 , DAC_vs_mock; GSM107603 , TSA_VS_mock; GSM107604 , DNMT1_vs_WT; GSM107605 , DNMT3B_vs_WT; GSM107606 , DKO_vs_WT; GSM107607 , WT_vs_DKO; GSM107660 , DAC_vs_mock_2; GSM107662 , TSA_vs_mock_2; GSM107663 , WT_vs_DNMT1; and GSM107664 , WT_vs_DNMT3B . The GEO series in which all ten arrays are linked may be found under accession number GSE4763 .
|
Loss of gene expression in association with aberrant accumulation of 5-methylcytosine in gene promoter CpG islands is a common feature of human cancer . Here , we describe a method to discover these genes that permits identification of hundreds of novel candidate cancer genes in any cancer cell line . We now estimate that as much as 5% of colon cancer genes may harbor aberrant gene hypermethylation and we term these the cancer “promoter CpG island DNA hypermethylome . ” Multiple mutated genes recently identified via cancer resequencing efforts are shown to be within this hypermethylome and to be more likely to undergo epigenetic inactivation than genetic alteration . Our approach allows derivation of new potential tumor biomarkers and potential pathways for therapeutic intervention . Importantly , our findings illustrate that efforts aimed at complete identification of the human cancer genome should include analyses of epigenetic , as well as genetic , changes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"genetics",
"and",
"genomics",
"homo",
"(human)"
] |
2007
|
Comparing the DNA Hypermethylome with Gene Mutations in Human Colorectal Cancer
|
Hantaan virus ( HTNV ) , a negative sense tripartite RNA virus of the Family Bunyaviridae , is the most prevalent hantavirus in the Republic of Korea ( ROK ) . It is the causative agent of Hemorrhagic Fever with Renal Syndrome ( HFRS ) in humans and maintained in the striped field mouse , Apodemus agrarius , the primary zoonotic host . Clinical HFRS cases have been reported commonly in HFRS-endemic areas of Gyeonggi province . Recently , the death of a member of the ROK military from Gangwon province due to HFRS prompted an investigation of the epidemiology and distribution of hantaviruses in Gangwon and Gyeonggi provinces that border the demilitarized zone separating North and South Korea . To elucidate the geographic distribution and molecular diversity of HTNV , whole genome sequences of HTNV Large ( L ) , Medium ( M ) , and Small ( S ) segments were acquired from lung tissues of A . agrarius captured from 2003–2014 . Consistent with the clinical incidence of HFRS established by the Korea Centers for Disease Control & Prevention ( KCDC ) , the prevalence of HTNV in naturally infected mice in Gangwon province was lower than for Gyeonggi province . Whole genomic sequences of 34 HTNV strains were identified and a phylogenetic analysis showed geographic diversity of the virus in the limited areas . Reassortment analysis first suggested an occurrence of genetic exchange of HTNV genomes in nature , ROK . This study is the first report to demonstrate the molecular prevalence of HTNV in Gangwon province . Whole genome sequencing of HTNV showed well-supported geographic lineages and the molecular diversity in the northern region of ROK due to a natural reassortment of HTNV genomes . These observations contribute to a better understanding of the genetic diversity and molecular evolution of hantaviruses . Also , the full-length of HTNV tripartite genomes will provide a database for phylogeographic analysis of spatial and temporal outbreaks of hantavirus infection .
Viruses in the Hantavirus genus of the family Bunyaviridae are negative-sense single-stranded RNA virus containing Large ( L ) , Medium ( M ) , and Small ( S ) segments [1] . Hantaviruses pose an emerging public health threat , with about 200 , 000 cases of human disease annually worldwide and fatality rates of 1–36% [2 , 3] . Old World hantaviruses , e . g . , Hantaan virus ( HTNV ) , Puumala virus ( PUUV ) , Seoul virus ( SEOV ) , and Dobrava-Belgrade virus ( DOBV ) , are etiologic agents of Hemorrhagic Fever with Renal Syndrome ( HFRS ) in Eurasia [4] . In America , Hantavirus Pulmonary Syndrome ( HPS ) results from infections with New World hantaviruses , e . g . , Sin Nombre virus ( SNV ) and Andes virus ( ANDV ) [5] . In humans , HFRS and HPS are contracted by inhaling aerosolized infectious particles from rodent salvia , urine , and feces [6] . Hantavirus infections are highly endemic and cause severe diseases in humans . However , there are no effective therapies or vaccines against these viruses . HTNV is harbored by striped field mice ( Apodemus agrarius ) , which constitute about 70% of the wild rodent population in the Republic of Korea ( ROK ) [7] . HFRS incidences increase during the spring and fall , due to the dynamics of rodent populations [8] . Observance of HFRS cases for military personnel and civilians near the demilitarized zone ( DMZ ) led to an investigation of the geographic distribution and molecular epidemiology of HTNV in Gangwon and Gyeonggi provinces [9–11] . Over decades , our studies have demonstrated the molecular similarities and diversity of hantaviruses using viral genomic sequences identified from HFRS patients and rodents where the patients were exposed in Gyeonggi province , the highest endemic area in the ROK [8 , 9 , 12–15] . However , the molecular prevalence of hantaviruses in Gangwon province remains unknown . Reassortment is a genetic event that results in the exchange of genome segments , and it is a major molecular mechanism to confer genetic diversity [16] . Influenza virus , a negative-sense segmented RNA virus , frequently generates reassortants by switching genomes of viruses from different hosts . Reassortment can play an important role in viral fitness , transmission , and pathogenesis of segmented RNA viruses [17] . The genetic reassortment of hantaviruses has been reported naturally and experimentally [18 , 19] . A natural reassortment of SNV within deer mice ( Peromyscus maniculatus ) occurred with the exchange of the M segment . Genetic exchange of the M segment of DOBV was observed between low pathogenic DOBV-Aa and highly pathogenic DOBV-Af in vitro . In addition , HTNV strains appeared to be highly divergent in the limited region of Guizhou in China , with the generation of S segment reassortants [20] . In this study , A . agrarius was collected in Gangwon and Gyeonggi provinces from 2003–2014 . To investigate the molecular epidemiology and distribution of HTNV , serological and molecular screening of HTNV was performed from lung tissues of the rodents . Using 34 of complete sequences of HTNV tripartite genomes , phylogenetic analyses show well-supported geographic clusters of the L , M , and S segments in the ROK . Reassortment analysis first demonstrated that HTNV in Dagmar North ( DN ) , Paju , consists of the heterogeneous L segment and homogeneous M and S segments , suggesting that the reassortment of HTNV may occur in nature . This study provides a better understanding of the genetic diversity and molecular evolution of HTNV tripartite genomes in the ROK . The whole genome sequences of HTNV will establish a database for the phylogeographic analysis and surveillance of endemic outbreaks of hantavirus infection .
Trapping of rodents was approved by US Forces Korea ( USFK ) in accordance with USFK Regulation 40–1 “Prevention , Surveillance , and Treatment of Hemorrhagic Fever with Renal Syndrome” . Wild rodents were euthanized by cardiac puncture and tissues were dissected under isoflurane anesthesia . All procedures and handling of rodents were conducted under an approved protocol by the Korea University Institutional Animal Care and Use Committee ( KUIACUC , #2010–212 ) . Rodents were captured in Gangwon and Gyeonggi provinces of the ROK , from 2003–2014 using live-capture Sherman traps ( 7 . 7 cm by 9 cm by 23 cm; H . B . Sherman , USA ) . A total of 100 traps were placed at intervals of about 4 m to 5 m at various US military training areas and ROK civilian sites in Cheorwon , Chuncheon , Hwacheon , Inje , Pyeongchang , Yanggu , and Yangyang in Gangwon province , and Dongducheon , Paju , Pocheon , Phyeongtaek , and Yeoncheon in Gyeonggi province each day over 2–3 consecutive days for each trapping period ( Fig 1 ) . Rodent sera , diluted 1:32 , were placed into duplicate acetone-fixed wells of Vero E6 cells infected with HTNV , and the wells incubated at 37°C for 30 min . After incubation , the slides were washed with Phosphate-Buffered Saline ( PBS ) and then fluorescein isothiocyanate ( FITC ) -conjugated anti-mouse IgG ( ICN Pharmaceuticals , Inc . , USA ) added to each well and incubated at 37°C for 30 min . Following washes , the slides were examined for virus-specific fluorescence , using an Axioscope fluorescent microscope ( Carl Zeiss AG , Germany ) . Lung tissues were mechanically homogenized using a FastPrep-24 5G Instrument ( MP Biomedicals , USA ) with TRI Reagent Solution ( Ambion , USA ) . Total RNA was extracted from lung tissues using a Hybrid R Kit ( GeneAll , Korea ) according to the manufacturer’s specifications . cDNA was synthesized using M-MLV ( Promega , USA ) with random hexamers or OSM55 ( 5'-TAGTAGTAGACTCC-3' ) [21] . First and nested PCR were performed in 25 μl reaction mixtures containing 200 μM dNTP ( Elpis Biotech , USA ) , 0 . 25 U of Super-Therm Taq DNA polymerase ( JMR Holdings , UK ) , 1 . 5 μl of DNA , and 5 pM of each primer . Oligonucleotide primer sequences for the first and nested PCR were specifically designed for HTNV . For the first and nested PCR , the initial denaturation was performed at 94°C for 4 min , followed by six cycles of denaturation at 94°C for 30 sec , annealing at 37°C for 30 sec , elongation at 72°C for 1 min , followed by 32 cycles of denaturation at 94°C for 30 sec , annealing at 42°C for 30 sec , elongation at 72°C for 1 min , and then elongation at 72°C for 5 min . PCR products were extracted using a PCR Purification Kit ( Cosmo GENETECH , Korea ) , and DNA sequencing performed in an Automatic Sequencer , Model ABI 3730XL DNA Analyzer ( Applied Biosystems , USA ) . Viral cDNA was synthesized with random hexamers or OSM55 . PCR was performed using specific primer sets covering the whole tripartite genome of HTNV . The PCR program was as follows: a cycle of 95°C for 5 min , 6 cycles of 95°C for 15 sec , 35°C for 30 sec , and 72°C for 20 sec , 30 cycles of 95°C for 15 sec , 42°C for 30 sec , and 72°C for 20 sec and then a cycle of 72°C for 3 min . The nucleotide sequences of HTNV L , M , and S segments were determined from virus-infected lungs of A . agrarius . Sequences were aligned and compared with HTNV sequences available in GenBank , using the ClustalW tool in the Lasergene program , version 5 ( DNASTAR , USA ) . For the phylogenetic analysis , the Neighbor-Joining ( NJ ) , Maximum Likelihood ( ML ) , and Bayesian methods ( MrBayes 3 . 2 . 2 Program ) were used . Topologies were evaluated by a bootstrap analysis of 1000 iterations . Alignments of HTNV sequences were analyzed using RDP , GENECONV , MAXCHI , CHIMAERA , 3SEQ , BOOTSCAN , and SISCAN in the Recombination Detection Program 4 ( RDP4 ) [22] , with concatenated L , M , and S segments . P-values under 0 . 05 were considered statistically significant . All parameters were left at the default RDP settings . The whole genome sequences of HTNV from reassortants , parents , and in- and out-groups were aligned and used to construct maximum likelihood trees of the individual segment in MEGA 5 . 2 [23] .
From 2003–2014 , a total of 5 , 929 striped field mice , A . agrarius , were collected in Cheorwon , Chuncheon , Hwacheon , Inje , Pyeongchang , Yanggu , and Yangyang in Gangwon province , and Dongducheon , Paju , Pocheon , Phyeongtaek , and Yeoncheon in Gyeonggi province ( Table 1 ) . The geographic distribution of rodent trapping sites included military training sites [Dagmar North ( DN ) , Twin Bridge Training Area North ( TBTA-N ) , and Twin Bridge Training Area South ( TBTA-S ) in Paju; Nightmare Range ( NR ) in Pocheon; Fire Point 131 ( FP131 ) in Yeoncheon] ( Fig 1 ) . To examine the positivity of anti-HTNV IgG , immunofluorescence antibody ( IFA ) test was performed using sera collected from A . agrarius . A total of 774/5 , 929 ( 13 . 1% ) A . agrarius were seropositive for anti-HTNV IgG . Among the 774 samples , 36 ( 4 . 7% ) and 738 ( 95 . 3% ) of the rodents were captured in Gangwon and Gyeonggi provinces , respectively . Fig 2 showed the number of A . agrarius which was positive for anti-HTNV IgG from trapping sites . Anti-HTNV IgG in A . agrarius from Chuncheon and Dongducheon was not detected . Total RNA was extracted from the lung tissues of seropositive samples and determined for the presence of HTNV RNA by RT-PCR . A 393-nucleotide partial sequence of HTNV M segment was recovered from 544/774 ( 70 . 3% ) of the seropositive samples; including 18 ( 3 . 3% ) and 526 ( 96 . 7% ) A . agrarius in Gangwon and Gyeonggi provinces , respectively ( Fig 2 ) . To obtain the whole tripartite genome sequences of HTNV , conventional PCR was performed using multiple primer sets . The whole genome sequencing of the HTNV L , M , and S segments was accomplished for 34 HTNV strains , including 3 strains from Cheorwon , 2 strains from Hwacheon , 6 strains from Yanggu , 6 strains from DN in Paju , 8 strains from TBTA-N and -S in Paju , 3 strains from NR in Pocheon , and 6 strains from FP131 in Yeoncheon ( Table 2 ) . The 5´ and 3´ end sequences of HTNV L , M , and S segments were determined by Rapid Amplification of cDNA Ends ( RACE ) experiments . The full-length sequences of HTNV tripartite RNA genomes were phylogenetically analyzed by Neighbor-Joining ( NJ ) , Maximum Likelihood ( ML ) , and Bayesian methods . The genomic sequences of HTNV L , M , and S segments formed geographic clusters , respectively ( Figs 3–5 ) . For the HTNV L segment , the L4 group ( Cheorwon ) phylogenetically grouped closely with the L5 group ( Pocheon ) . The L8 group ( Yanggu ) formed a group that was distinct from all other strains from Gyeonggi province . The L6 group ( Hwacheon ) formed a phylogenetic cluster with the L7 group ( DN , Paju ) . Phylogenetic analyses of the HTNV M and S segments demonstrated that the M4 and S4 groups ( Cheorwon ) clustered with the M5 and S5 groups ( Pocheon ) . The M and S segments of HTNV strains from Yanggu ( M8 and S8 ) and Hwacheon ( M7 and S7 ) in Gangwon province formed a geographic and distinct clusters from strains obtained in Gyeonggi province . The percent of nucleotide and amino acid sequence homologies of HTNV newly acquired was shown in S1–S3 Tables . Phylogenetic analyses showed that the 7 group ( DN , Paju ) contained heterogeneous L segment which was clustered with the L6 group ( Hwacheon ) ( Fig 3 ) . However , the M7 and S7 segments of HTNV formed homogeneous clusters with the M2 and S2 groups ( TBTA-N , Paju ) , respectively ( Figs 4 and 5 ) . To evaluate the possibility of a reassortment event in the HTNV genomes , the concatenated HTNV tripartite genomes from Gangwon and Gyeonggi provinces were aligned and analyzed using RDP4 software package . Fig 6A shows P-values ranging from 1 . 632E-02 to 1 . 114E-17 based on the various recombination/reassortment analysis programs . The RDP recombination consensus score ( RDPRCS ) of the 7 group ( DN , Paju ) is 0 . 522 . Fig 6B shows high percent of bootstrap support for a grouping of the HTNV M and S segments from the 7 group ( DN , Paju ) and the 2 group ( TBTA-N , Paju ) , whereas the L segment shows a high similarity between the 7 group ( DN , Paju ) and the 6 group ( Hwacheon ) . Eight HTNV stains representing reassortants , parents , and in- and out- groups were used to generate maximum likelihood trees of the HTNV L , M , and S segments , respectively ( Fig 6C ) . These results suggest that the 7 group ( DN , Paju ) were reassortants in nature based on the heterogeneity of the L segment , as the L segment formed a lineage with the 6 group ( Hwacheon ) in Gangwon province , while the M and S segments clustered with the 2 group ( TBTA-N , Paju ) in Gyeonggi province .
HFRS is highly endemic in Gangwon and Gyeonggi provinces , ROK , affecting both military personnel and civilians . For decades , we conducted epidemiological and phylogeographic analyses of HTNV in Gyeonggi province . Recent clinical cases of HFRS in Gangwon province , including one deceased patient in 2013 , prompted us to investigate the geographic distribution and diversity of HTNV in Gangwon province [24] . A total of 5 , 929 A . agrarius were collected in the endemic areas of Gangwon and Gyeonggi provinces from 2003–2014 . IFA tests showed that 774 ( 13 . 1% ) A . agrarius were seropositive for anti-HTNV IgG . HTNV-specific RT-PCR showed that 544/774 ( 70 . 3% ) A . agrarius were positive for a partial HTNV M segment . Based on the serological and molecular tests , there was a lower prevalence of HTNV infections among A . agrarius in Gangwon province than in Gyeonggi province . Although the number of A . agrarius was much lower for Gangwon province , these data suggest that Gyeonggi province poses significantly higher HFRS health risks than Gangwon province [8 , 9 , 14 , 15] . To support this , the incidence of HFRS patients in Gangwon province was lower than that in Gyeonggi province , demonstrating a correlation between rodent seroprevalence and human disease data [25] . Using the complete genomic sequences of 34 HTNV strains newly obtained from eight different trapping sites , the phylogenetic trees of HTNV L , M , and S segments demonstrated well-supported geographic clusters and showed the genetic diversity in the restricted areas . Diversification of HTNV , the Old World hantavirus , has been observed in China , a high HFRS-endemic area [20] . In Guizhou , high level of the molecular diversity of HTNV strains were observed . The ADNV , a New World hantavirus from southern South America , also exhibited high molecular diversity of the M segment , resulting in five different lineages based on their geographic origins in Argentina and Chile [26 , 27] . Consistent with these observations , HTNV strains in the restricted areas of the ROK showed a high molecular diversity representing geographic distinct clusters . Reassortment , recombination , and genetic drift are molecular genetic mechanisms that confer the genetic diversity in RNA viruses in nature [28] . Segmented RNA viruses preferentially give rise to genetic reassortment rather than recombination . A reassortment event predicted by RDP4 was considered significant if it satisfied at least 2 criteria with a P-value ( p ) <0 . 05 and the RDPRCS was >0 . 6 [29] . When P <0 . 05 and the RDPRCS was between 0 . 4 and 0 . 6 , the genetic event was considered possible . An RDPRCS under 0 . 4 with P <0 . 05 was a cause to reject the genetic event . The HTNV strains ( the 7 group ) from DN were likely reassortants since these strains showed P <0 . 05 and an RDPRCS of 0 . 522 ( HTNV from Hwacheon , 0 . 322; HTNV from TBTA-N , 0 . 156 ) . The genetic reassortment of bunyaviruses in nature has been previously reported [30 , 31] . Reassortments of SNV have more commonly been described in M segments than S or L segments in nature and in vitro [18 , 31] . The L , M , and S segments of hantaviruses encode the viral RNA-dependent RNA-polymerase ( the L protein ) , two surface glycoproteins ( Gn and Gc ) , and the nucleocapsid ( N ) , respectively [32] . The less common reassortants of L or S segments than M segment might be associated with the function of these viral proteins . The maintenance of L and S segments might be beneficial for replication , transcription , and assembly , resulting in the production of the progeny possessed appropriate viral fitness . Still , given our limited understanding of this genetic mechanism , the existence of reassortants that contain different combinations of segments should not be ignored . Infection with Guaroa virus generated the L segment reassortant , whereas the S segment reassortant was observed between Bunyamwera and California encephalitis viruses [33] . In this study , the reassortment analysis demonstrated that HTNV ( the 7 group ) from DN in Paju showed heterogeneous L segment , but homogeneous M and S segments . Whether the exchange of L segment is a determinant of the fitness and pathogenesis of HTNV remains to be studied . In conclusion , we report the phylogeographic analysis of full-length HTNV sequences from A . agrarius collected in HFRS-endemic areas of Gangwon and Gyeonggi provinces , ROK . These results demonstrate the geographic diversity and a possible reassortment of HTNV in nature . This study greatly increases our understanding of the genetic diversity and molecular evolution of HTNV in the hantavirus-endemic areas . The whole sequences of HTNV tripartite genomes will provide a database for the phylogeographic analysis and surveillance of hantavirus infection from HFRS patients and natural reservoir hosts .
|
Hemorrhagic Fever with Renal Syndrome ( HFRS ) and Hantavirus Pulmonary Syndrome ( HPS ) are endemic zoonotic infectious diseases caused by hantaviruses that belong to the Family Bunyaviridae containing negative-sense tripartite RNA genomes . Hantaviruses pose a critical emerging public health threat , with up to 200 , 000 clinical cases reported annually worldwide with 1–36% case fatality rates . In humans , hantavirus-borne diseases are contracted by the inhalation of viruses aerosolized from rodent excreta . However , there is no effective therapeutic or vaccine to prevent from the disease . Whole genome sequences of Hantaan virus ( HTNV ) were acquired from lung tissues of Apodemus agrarius captured in HFRS-endemic areas of the Republic of Korea ( ROK ) . Phylogenetic analyses demonstrated that sequences of the HTNV tripartite genomes clustered geographically , showing broad diversity of HTNV throughout the areas surveyed . Reassortment analysis first suggested a natural occurrence of the HTNV genetic exchange in the ROK . These observations contribute to a better understanding of the genetic diversity and molecular evolution of hantaviruses in HFRS-endemic regions . The complete sequences of HTNV genomes will provide a database for the phylogeographic analysis and surveillance of endemic hantavirus-borne diseases .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"sequencing",
"techniques",
"biogeography",
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"population",
"genetics",
"vertebrates",
"animals",
"mammals",
"hemorrhagic",
"fever",
"with",
"renal",
"syndrome",
"phylogenetic",
"analysis",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"population",
"biology",
"mammalian",
"genomics",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"infectious",
"diseases",
"geography",
"genomics",
"biological",
"databases",
"phylogeography",
"molecular",
"biology",
"animal",
"genomics",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"rodents",
"dna",
"sequence",
"analysis",
"earth",
"sciences",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"computational",
"biology",
"evolutionary",
"biology",
"amniotes",
"genomic",
"databases",
"organisms"
] |
2016
|
Genetic Diversity and Reassortment of Hantaan Virus Tripartite RNA Genomes in Nature, the Republic of Korea
|
While Grenada attained a zero-human-rabies case status since 1970 , the authors conducted the first study to assess knowledge , attitudes , and practices that may contribute to this status as well as to receive feedback on the rabies control program in Grenada . A cross-sectional survey was conducted in July , 2017 with 996 households on the mainland . A questionnaire was administered to collect information on knowledge of rabies and prevention , vaccination practices , perception of institutional responsibilities for rabies control , and evaluation of the anti-rabies program . Of the 996 households , 617 ( 62% ) had owners of animals that can be infected with rabies and were included in the analysis . Respondents were very aware of rabies as a disease that can infect animals and humans . The rate of participation in the vaccination program was 51 . 6% for pets and 38 . 0% for livestock . About 40% of respondents were knowledgeable about the extent of protection from the rabies vaccine . Respondents did not demonstrate exceptionally high levels of knowledge about animals that were likely to be infected with rabies , neither the anti-rabies programs that were conducted in Grenada . The three most frequent recommendations made to improve the rabies-control programs were: increase education programs , control the mongoose population , and expand the vaccination period each year . Conducting a comprehensive national rabies education program , expanding the vaccination program , and increasing the rate of animal vaccination are important steps that need to be taken to maintain the current zero-human-case status .
Rabies is a zoonotic disease that is endemic in many countries , including in the Caribbean region [1] . The World Health Organization ( WHO ) reported that the disease kills tens of thousands of people every year [2] . Most deaths occur in developing countries and rural communities [3 , 4] . Global Alliance for Rabies Control ( GARC ) estimates approximately 55 , 000 people worldwide , mostly children , become infected and die annually [5] . Most human cases are as a result of bites from infected dogs . Bat , fox , and mongoose are also common hosts for rabies [6] . Notification of human cases of the disease declined in Latin America and the Caribbean from about 250 cases in 1990 to less than 10 cases in 2010 due to the implementation of dog rabies control programs [4] . Nonetheless , the high mortality rate among animal and human cases across the globe emphasizes the need for active surveillance and prevention programs [2] . Since 1983 , the Pan American Health Organization ( PAHO ) has been providing technical support for countries to eliminate rabies [1] . In April , 2015 , the first Regional Conference on Research and Surveillance on Emerging and Vector-Borne Animal Diseases in the Caribbean was held in Guadeloupe with a primary focus on rabies in the Caribbean [6] . Information was presented on the prevalence of rabies and prevention programs in the respective countries . The highest number of rabid animal cases per year was reported in Trinidad ( 6–10 cases ) and Puerto Rico ( >20 cases ) , while 1–5 cases per year were reported in Grenada , Belize , and Guyana , respectively [6] . Surveillance information was not provided for Suriname , Cuba , Haiti , and the Dominican Republic [6] . Review of the literature shows that few studies were published on knowledge , attitudes , and practices ( KAP ) relating to rabies in Caribbean countries . References were found for several studies in other regions , such as in African countries [7–9] and in the Americas [1 , 10–12] . In Grenada , the Ministry of Health ( MOH ) conducts several activities aimed at controlling the spread of rabies in animals and to maintain zero-human-case status . The programs conducted by the Ministry includes anti-rabies vaccination , stray dog control , mongoose trapping , public education , investigating reports of persons bitten by animals , and rabies surveillance . To date , no study was conducted to assess knowledge , attitudes , and practices that may contribute to maintain zero-human-case status or to assess risk factors for animal-human transmission of the disease in Grenada . The objective of this study , therefore , is to assess knowledge , attitudes , and practices in Grenada regarding rabies and to receive feedback from the public about the rabies control program in the MOH . The study was supported by PAHO , which is the sub-regional organization/body of WHO to provide support for countries in Latin America and the Caribbean to eliminate rabies . Based on the objective of the study , information was collected on the public’s response to vaccination programs , animal vaccination coverage , perceptions about institutional responsibilities for anti-rabies programs , evaluation of the Ministry’s anti-rabies programs , and knowledge about rabies and prevention . The findings can be used to guide the MOH in expanding and enhancing its anti-rabies programs to reduce the risk of the disease in animals and humans .
Ethical approval for the study was granted by the St . George’s University ( SGU ) Institutional Review Board ( IRB ) . The study design was a cross-sectional survey , administered to households in all parishes on the mainland in July , 2017 . The State of Grenada includes the mainland , Grenada , and two smaller dependency islands , Carriacou and Petite Martinique . The mainland is divided into six parishes: St . Andrew , St . Patrick , St . Mark , St . John , St . George , and St . David . Rabies is endemic on the mainland , but not in Carriacou and Petite Martinique . Therefore , Carriacou and Petite Martinique were excluded from the study . Using a hypothesized frequency of outcome at 50% , confidence limit of 5% , and the total of 33 , 670 households as determined by the Central Statistics Office ( CSO ) in Grenada , a minimum sample size of 570 was calculated . Given that a reference was not available of households with animals in Grenada , a total of 1000 households were decided for contact to ensure that at least 570 households were identified for inclusion in the study . A multi-stage cluster sampling strategy was used with Enumeration District ( ED ) as the primary sampling units . In the first stage of sampling , the ED was randomly selected within the respective parish . Of the 283 EDs on the mainland , a total of 62 ED were randomly selected for inclusion in the study . In the second stage , households were randomly selected within the ED . All households in the ED on the mainland that were randomly selected for inclusion in the study were contacted . At the beginning of the survey , residents were asked whether anyone in the household owned pets and/or livestock ( not including fishes , turtles , birds , snakes ) , which were considered as susceptible to rabies–that is , the animal can be infected with the rabies virus . The questions on KAP regarding rabies were only administered in households with owners of pets and livestock that were susceptible to rabies . The oldest person , above 18 years , who owned animals ( s ) in the household , was selected for the interview . If the owner was under 18 years , the head of the household , that was 18 years or older , was selected . One survey was administered per household in a face-to-face interview . Written consent was required for participation in the survey . The survey was administered during the first two weeks in July , 2017 . The questionnaire included primarily closed-ended questions and a few open-ended questions focusing on four main areas: participation in vaccination programs; perceptions about institutional responsibilities; evaluation of anti-rabies programs; and knowledge about rabies and prevention . A total of 46 questions were included and the responses were filled on the questionnaire by the interviewer . Each participant was assigned a numeric code . Data were entered and analyzed using IBM SPSS software ( V . 24 ) . Descriptive and inferential analyses were performed to determine: Chi square analysis was conducted to investigate relationships between demographic characteristics–education , gender , age group—of the respondents and KAP regarding rabies . Results that were statistically significant at alpha 0 . 05 are reported .
Residents in a total of 996 households responded to the survey of which 617 ( 61 . 9% ) households had owners of animals that were susceptible to rabies ( animals that can be infected with the virus ) , distributed as follow: 271 ( 43 . 9% ) in St . George , 128 ( 20 . 7% ) in St . Andrew , 92 ( 14 . 9% ) in St . David , 90 ( 14 . 6% ) in St . Patrick , 27 ( 4 . 4% ) in St . John , and 9 ( 1 . 5% ) in St . Mark . Fig 1 shows the distribution of respondents across the parishes on the mainland in Grenada . There were about equal proportion of males ( 307 , 49 . 8% ) and females ( 299 , 48 . 5% ) in the study . Representation was also fairly consistent across age groups except for a slightly higher percentage of respondents aged 46–55 years . Most of the respondents completed school at the primary ( 261 , 42 . 3% ) and secondary level ( 179 , 29 . 0% ) . Table 1 shows the demographic profile of the respondents . The majority of households owned dogs , 323 ( 52 . 3% ) , while 86 ( 14 . 0% ) of the households each had owners with cats , sheep and goats . Of the 617 respondents , 602 ( 97 . 6% ) reported they had heard of rabies . School/work and electronic media ( radio , television , internet/social media ) were the primary sources of learning about rabies . In each case , more than 50% of respondents correctly identified dogs , mongooses , cats , and sheep/goats , as susceptible to rabies while 19 ( 3 . 1% ) respondents stated they were not sure which of the animals were susceptible to the disease . Monkey , pig , bat , donkey , and cattle were correctly identified as animals that are susceptible to the disease by 30% or fewer respondents . Males were more likely to correctly identify sheep/goat χ2 ( 2 , N = 617 ) = 8 . 65 , p = . 013 ) , pig χ2 ( 2 , N = 617 ) = 10 . 75 , p < . 01 ) , monkey χ2 ( 2 , N = 617 ) = 18 . 22 , p < . 01 ) , donkey/horse χ2 ( 2 , N = 617 ) = 10 . 24 , p < . 01 ) , mongoose χ2 ( 2 , N = 617 ) = 22 . 93 , p < . 01 ) , bat χ2 ( 2 , N = 617 ) = 13 . 94 , p < . 01 ) , cattle χ2 ( 2 , N = 617 ) = 31 . 43 , p < . 01 ) as animals that were susceptible to the disease . Table 2 shows the number and percentage of respondents that correctly identified animals that are susceptible to rabies . A few respondents , 23 ( 3 . 7% ) , stated they did not know how rabies was transmitted while 41 ( 6 . 6% ) incorrectly stated the disease was transmitted through insect bites , 35 ( 5 . 7% ) stated the disease can be transmitted through contact with an infected person , and 7 ( 1 . 1% ) stated the disease can be transmitted through sneezing . The majority , 528 ( 85 . 6% ) , correctly stated that the disease can be transmitted by animal bite . Apart from identifying aggression as a sign of rabies in animals , less than 40% in each case correctly identified any of the other signs . Males were more likely to correctly identify aggression χ2 ( 2 , N = 617 ) = 13 . 27 , p = . 02 ) , not afraid of people χ2 ( 2 , N = 617 ) = 12 . 01 , p < . 01 ) , making unusual sounds/howling , bawling , bellowing χ2 ( 2 , N = 617 ) = 6 . 188 , p = . 05 ) the signs of rabies in animals . Table 3 shows the number and percentage of respondents that correctly identified signs of rabies in animals . About two-third of respondents , 462 ( 74 . 9% ) , correctly identified vaccination to prevent transmission of rabies . A small percentage of respondents , 15 ( 2 . 4% ) , felt that rabies was not preventable , and 72 ( 12% ) did not know of any way to prevent the disease . Respondents also demonstrated little knowledge about the protection that was provided for animals from vaccination . A total of 368 ( 62 . 2% ) respondents stated they did not know how for long the vaccine protected the animals . Only 79 ( 13 . 3% ) respondents correctly stated 1 year . Fig 2 shows ( 162 ) 51 . 6% of respondents who ever vaccinated their pets did so in the past year based on the requirement of the MOH for annual vaccination . Respondents who completed education at college and university levels were less likely to ever vaccinate their pet , χ2 ( 2 , N = 515 ) = 9 . 837 , p < . 01 ) . The percentage of respondents that vaccinated pets in the last year did not differ by education , gender , or age group . Fig 3 shows that less than half of the respondents reported they had vaccinated livestock at any time . Fig 4 shows that slightly above one-third of respondents reported they vaccinated livestock in the last year . The percentage of respondents that vaccinated livestock in the last year differ by education . Respondents who completed education at college and university levels were less likely to vaccinate livestock at any time compared to respondents who completed education at lower education levels χ2 ( 2 , N = 100 ) = 10 . 07 , p < . 01 ) ) . Overall , only 168 ( 27 . 2% ) respondents reported they had vaccinated animals in the government’s anti-rabies program . Of 314 respondents who reported vaccinating pets at any time , 261 ( 83 . 1% ) reported they vaccinated dogs , 28 ( 8 . 9% ) vaccinated cats , and 6 ( 1 . 9% ) vaccinated other pets . Of 189 respondents who reported vaccinating livestock at any time , 44 ( 23 . 3% ) vaccinating sheep or goat , respectively , 9 ( 4 . 8% ) vaccinated pigs , and 18 ( 9 . 5% ) vaccinated cattle . Several reasons were given by the respondents ( 118 , 62 . 2% ) for failing to vaccinate livestock; most commonly , respondents did not know about the Government’s anti-rabies vaccination program ( 48 , 40 . 7% ) , vaccination took too long ( 44 , 37 . 3% ) , transportation problems to bring the animals to the vaccination site ( 13 , 11 . 0% ) , and not being at home when the vaccination team was in the area ( 11 , 9 . 3% ) . Almost half of the respondents , 281 ( 45 . 5% ) , also stated that the MOH should remain in charge of the anti-rabies program , 84 ( 13 . 6% ) felt that animal owners should control the program , and 47 ( 7 . 6% ) respondents felt the Ministry of Agriculture should be in charge . Apart from the vaccination program , generally , less than 20% of respondents knew about the other programs , except for about one third that was also aware of the mongoose trapping and stray dog control programs . Table 4 shows respondents’ awareness of the anti-rabies programs in the MOH . The MOH’s vaccination team was mostly rated as helpful , friendly , informative , and knowledgeable . Table 5 shows the respondents evaluation of the anti-rabies vaccination team in the Ministry of Health . Increasing education programs , controlling the mongoose population , and increasing the number of times that the vaccination program is conducted in each year were the most frequent recommendations made to improve the rabies control program . To disseminate information about the anti-rabies program , respondents most commonly suggested using the MOH’s public address system in communities ( 82 , 13 . 3% ) and radio announcements ( 72 , 11 . 7% ) .
This is the first study conducted in Grenada to assess KAP regarding rabies and to assess the public’s feedback on the MOH’s anti-rabies programs . While this is the first study to assess the rabies control program in Grenada , the findings from other studies and clinical data indicate that the Grenadian population is at risk for rabies . A study conducted in Grenada from 2011–2013 on rabies prevalence in mongooses that were trapped found that about 0 . 5–1 . 5% of the animals were infected with the rabies virus [13] . Another study shows that of 173 animals tested between 2001–2016 , 64 ( 36 . 4% ) tested positive for the rabies virus with the highest prevalence in dogs and mongooses [14] . Both dogs and mongooses are likely to come in contact with humans and , therefore , pose a risk of transmission of the virus to humans through bites from infected animals . In 2015 , a total of 384 cases were reported to the MOH of persons bitten by dogs ( 314 cases reported by community clinics , 70 cases reported by hospitals ) [15] . As such , this study provides information that can be used to guide the MOH and its partner institutions in enhancing the efforts to control rabies in Grenada . The data can also be used as a baseline for comparison with future studies to determine whether there were changes in knowledge and behaviors relating to rabies . Other countries in the Caribbean region may refer to these findings to identify at-risk populations and to guide in the development of strategies to break the transmission cycle . Prior to conducting this study , the proportion of households with animals that are susceptible to rabies–that is , animals that can be infected with the rabies virus—was not established in Grenada . The findings of this study show that about 62% of households have animals that are susceptible to the disease . This percentage can be used as a reference for the scope that should be covered by the MOH’s anti-rabies program and to inform planning and resource allocation . Additionally , this reference can be used to plan sampling for other studies , such as , follow up studies to monitor changes in KAP among animal owners over time or immediately following interventions . During the time of this study , the government operated laboratory , with capacity to analyze samples from animals , was not functional . However , the services were provided at St . George’s University veterinary laboratory . This restricts capacity to handle post-exposure emergencies , particularly , by animal owners with limited resources to pay for the services at the University . Meanwhile , however , the MOH continues to purchase and administer rabies vaccine to individuals who are believed to be at risk after being bitten by animals . As mentioned above , this measure is not cost-effective . Efforts should be made to repair the laboratory in the shortest time period . In keeping with the Dog Control and Regulation Act of 2002 , registration of dogs and vaccination of domestic animals are the two approaches that were instituted by the MOH to manage disease transmission [14] . A study conducted by Keku et al . ( 2016 ) on stray dogs in Grenada found that more stray dogs were being captured and that the rate of dog registration and vaccination had decreased significantly between 2008–2012 [16] . As such , the provisions of the Control and Regulation Act of 2002 should be fully exercised , requiring dogs to be registered [14] and , thus , providing an avenue for better coordination and monitoring of the reach of the vaccination program . Free roaming dogs are at high risk for contracting rabies from wild animals , such as mongoose , and can transmit the disease to humans [16] . Controlling the population of free-roaming dogs is , therefore , a critical step in breaking the transmission line . Registration of other animals , such as livestock , should also be considered for ease of mobilization for vaccination and to monitor coverage . The findings show that , generally , respondents were not very knowledgeable about animals that are susceptible to rabies . Dog was identified by the majority of respondents , however , only a few respondents identified bat and there was low responses in identifying most of the other animals , such as pig and cattle . Most respondents also had limited knowledge about the signs of rabies in animals . Males were more likely to correctly identify animals that are susceptible to rabies as well as the signs of rabies in animals . This finding may indicate differences in access to information about rabies or the influence of livelihood practices—males are generally more involved in animal husbandry . In any case , the MOH should conduct an evaluation of education programs to ensure that there is equity in opportunities to learn about the disease . Further studies can also be conducted to investigate factors that may influence knowledge about rabies among males and females . Hunters , farmers , forest rangers , and other groups that may readily come in contact with animals are at higher risk and should be targeted for education programs . Increasing the level of knowledge about signs of rabies can also lead to increased reporting to surveillance . There was also very low participation in the vaccination program , especially by respondents who owned livestock . Apart from the MOH , vaccination is also provided by private veterinarians and at the animal clinic at St . George’s University suggesting options for participating in vaccination programs . Only education was found to be associated with participation in the animal vaccination program . Respondents with higher levels of education–college and above—were also less likely to vaccinate animals as compared to respondents who completed education at lower institutions . This information is critical for the MOH to guide in targeting strategies for rabies education programs . Apart from the vaccination program , there was limited awareness about the other anti-rabies programs in the MOH . The results show that there was little awareness about the other anti-rabies programs conducted by the MOH , including stray dog control , mongoose trapping , public education , investigating reports of persons bitten by animals , and rabies surveillance . This finding is also interesting , given that the majority of respondents also stated that they would call the MOH if they were bitten by an animal or suspected that an animal had rabies . The findings may indicate that , despite calling the MOH , the public was still uncertain about the courses of action that may be taken . In addition , there may be a lack of awareness by some community Health professionals of the existing rabies treatment protocol . The MOH should incorporate information about all the anti-rabies programs in a comprehensive education campaign . This step can also contribute to increase confidence in the MOH’s strategies to address rabies in various ways . While vaccination is encouraged to protect the health of the public , there is no legal mandate for public compliance . As such , to address the low participation in animal vaccination programs , the MOH would need to develop and utilize strategies that are appealing to the public . The current rate of participation in vaccination is not sufficient to achieve herd immunity . The WHO cited that 70% of dogs in an area must be vaccinated to achieve heard immunity [17] . The MOH may consider establishing a committee to develop a strategic plan and oversee interventions to address gaps in knowledge and practices . Among the main reasons suggested for the low vaccination coverage were unawareness of the anti-rabies program and the long time taken to vaccinate animals . Challenge in transporting animals to vaccination sites was also mentioned . As such , the MOH may need to consider the practicality and feasibility of providing vaccination services on farms and other convenient locations . Consideration should also be given to extending the vaccination programs to weekends and evenings to accommodate employees that work during regular hours . A fee-for-service system may be considered to support the anti-rabies program . The MOH can also explore the possibility of utilizing Oral Rabies Vaccine ( ORV ) . ORV was used in the United States from the mid-1990s to date to prevent and control wildlife rabies [18] . Investigations may be conducted to determine how factors such as climatic conditions , animal species , territorial behavior patterns and physiological characteristics of animals may affect the suitability of this initiative for the Grenada setting . In many ways , the findings in this study were similar to results in other countries [5 , 7 , 14] . In studies conducted in 2013 and 2014–2015 on KAP related to rabies in Ethiopia , it was found that while most respondents were aware of rabies , the majority did not have accurate knowledge about how rabies was transmitted , signs of the disease and prevention and treatment measures [19 , 20] . Dogs and cats were most commonly identified as animals involved in transmission of the virus and the majority of respondents also correctly stated that the disease can be transmitted through bites from animals [19 , 20] . These findings in Ethiopia reflect similar knowledge of respondents in this study . In another study conducted in Tanzania in 2009–2010 among 5 , 141 respondents , a similar result to Grenada was also achieved with regard to the low proportion of respondents who vaccinated animals [9] . About half of the respondents , 51 . 0% , had vaccinated dogs and several gaps were found in knowledge and practices that were associated with socio-economic status [9] . While most of the respondents in this study indicated that their primary place of learning about rabies was school/work , the findings also show inconsistencies in their knowledge that may reflect on the content of the education programs . Some insights were provided in a study that was conducted in Bangladesh in 2014 with teenage students in two high schools [8] . The researchers found that there was a low level of knowledge about rabies and poor handling of pets by the students and that increased the risk of transmission of the disease [8] . Following a deliberate and well-planned education intervention , the level of knowledge and practices improved . The MOH can benefit from the Bangladesh experience , noting that similar result can be achieved from implementing a planned program in schools in Grenada . There were limitations in conducting this study in Grenada . The representation of male and females in the study were consistent with the census ( 2011 ) proportions , however , the overwhelming majority of respondents had completed education at primary and secondary school levels . This inconsistency may be as a result of the time of day that the survey was conducted . People with lower education may likely be unemployed and at home during regular working hours . For studies in the future , one approach that can be taken to reduce this issue is to designate specific hours for data collection . Some questions need to be revised to be more specific and to improve the quality of the data . For example , the option of “school/work” should be separated to give a clearer indication of the specific place where the respondent first learned about rabies . Prior to use in another study , the questionnaire should be reviewed and compared with the tools used in other countries . The questionnaire can then be refined and used as a standard tool across the region allowing for comparability and collation of findings .
|
About 62% of households in Grenada owned animals that are susceptible to rabies . Pets , particularly dogs , were more commonly owned than livestock . Males were more likely to correctly identify animals that are susceptible to rabies . The level of participation in the animal vaccination program was low , especially among owners of livestock . A lack of knowledge about the free anti-rabies vaccination service and the length of time that it took to vaccinate animals were the most common problems associated with the low vaccination rate . Increasing education programs , controlling the mongoose population , and increasing the vaccination times per year were the most frequent recommendations to improve the Ministry of Health rabies control programs . The anti-rabies program remains a critical step to maintain the current zero-human-case status in Grenada . There is a need to also expand the vaccination program and increase the rate of animal vaccination .
|
[
"Abstract",
"Introduction",
"Method",
"Results",
"Discussion"
] |
[
"livestock",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"education",
"pathogens",
"immunology",
"tropical",
"diseases",
"geographical",
"locations",
"vertebrates",
"microbiology",
"social",
"sciences",
"sociology",
"animals",
"mammals",
"dogs",
"north",
"america",
"preventive",
"medicine",
"viruses",
"rabies",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"vaccination",
"and",
"immunization",
"caribbean",
"veterinary",
"science",
"rabies",
"virus",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonoses",
"medical",
"microbiology",
"schools",
"microbial",
"pathogens",
"agriculture",
"people",
"and",
"places",
"lyssavirus",
"eukaryota",
"grenada",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"amniotes",
"organisms"
] |
2019
|
Knowledge, attitudes, and practices regarding rabies in Grenada
|
Mayaro virus ( MAYV ) is an emerging , mosquito-borne alphavirus that causes a dengue-like illness in many regions of South America , and which has the potential to urbanize . Because no specific treatment or vaccine is available for MAYV infection , we capitalized on an IRES-based approach to develop a live-attenuated MAYV vaccine candidate . Testing in infant , immunocompetent as well as interferon receptor-deficient mice demonstrated a high degree of attenuation , strong induction of neutralizing antibodies , and efficacy against lethal challenge . This vaccine strain was also unable to infect mosquito cells , a major safety feature for a live vaccine derived from a mosquito-borne virus . Further preclinical development of this vaccine candidate is warranted to protect against this important emerging disease .
Mayaro virus ( MAYV ) is an important and growing human health concern in the neotropics . First isolated in Mayaro county , Trinidad in 1954 , cases of Mayaro fever ( MAY ) have since been reported in 9 different countries in northern South America [1] . In addition , serological surveys suggest that MAYV has expanded into the Central American countries of Costa Rica , Guatemala , and Panama [2] . Typical presentations of MAY consist of an acute febrile illness accompanied by headache , retro-orbital pain , myalgia , vomiting , diarrhea , and rash [3] . However , the hallmark manifestation of MAY is arthralgia [4] , which is often severe and debilitating , and can persist for up to a year , with recurring relapses possible . The high incidence of dengue fever in the same areas in which MAYV circulates , and the similarity of the initial signs and symptoms , leads to the misdiagnosis and underreporting of MAY cases [5] , [6]; therefore , MAYV is typically neglected as an important cause of tropical diseases . For example , in several areas of northern South America approximately 1% of all febrile illness that is clinically similar to dengue is caused by MAYV [7] . MAYV is a zoonotic pathogen that circulates in an enzootic cycle involving Haemagogus spp . mosquitoes and as yet unidentified vertebrate hosts [3] . Although seropositivity has been detected in birds and rodents , non-human primates have consistently demonstrated the highest rates of antibodies , suggesting that they are the principal reservoir hosts . Infection of humans typically occurs in communities near humid tropical forests , and is often associated with logging or other forest activities [1] , [8]–[10] . However , as land use and demographic changes in South America lead to human populations expanding within regions of tropical forest , an increasingly higher percentage of the population may be at risk [11] . In addition , the demonstration that the urban mosquito , Aedes aegypti , can transmit MAYV after exposure to bloodmeals with titers approximating human viremia levels [5] , [12] raises the concern that the virus could emerge into an urban transmission cycle similar to that of its close relative , chikungunya virus ( CHIKV ) . MAYV belongs in the family Togaviridae , genus Alphavirus . Despite circulating exclusively in the New World , MAYV belongs genetically , antigenically [13] , [14] . The genome of MAYV is a single-stranded , positive sense RNA , approximately 11 . 45 Kb in length that encodes 4 nonstructural proteins ( nsP1-4 ) on the 5′ end and 3 structural proteins on the 3′ end , including the capsid and envelope glycoproteins , E1 and E2 ( Fig . 1 ) [13] , [15] . Genomic RNA includes 2 open reading frames ( ORFs ) ; the nonstructural polyprotein ORF is translated in a cap-dependent manner from genomic RNA , while the structural polyprotein ORF is translated from a subgenomic RNA transcript , which is also capped [16] , [17] . There is no licensed vaccine available for MAY , and current control strategies rely on reducing human exposure to potentially infected mosquito vectors . Only one attempt to generate a vaccine for MAYV infection is described in the literature [18] . Formalin inactivation of wild-type ( wt ) MAYV strain TRVL15537 was tested in immunocompetent CD-1 mice using a single vaccination . This vaccine was immunogenic , and some efficacy was demonstrated via passive transfer of immune mouse sera to infant mice , followed by lethal challenge . The ideal MAYV vaccine would produce rapid , long-term immunity after a single dose to rapidly control outbreaks , with a low risk of adverse side effects . The vaccine would also need to be cost effective for use in resource-poor parts of Latin America , and easy to produce . For a live-attenuated vaccine , which typically meets most of these criteria , mosquito-transmission incompetent would also be highly desirable for use in non-endemic locations . To produce such a vaccine , we employed an attenuation strategy involving an encephalomyocarditis virus ( EMCV ) internal ribosome entry site ( IRES ) , which has been successfully used for other alphavirus vaccines [19]–[24] . Replacement of the subgenomic promoter reduces expression of the structural proteins , which are now translated via the IRES from genomic RNA , and the inefficient recognition of the IRES by insect ribosomes results in a phenotype that is also incapable of replicating in mosquito cells [25] . For this study , we tested the efficacy of an IRES-based vaccine candidate for MAYV ( henceforth called MAYV/IRES ) , which was highly attenuated , efficacious , and safe when tested in murine models .
A full-length genomic cDNA clone was generated from MAYV strain CH using RT-PCR and standard cloning methods as described previously [20] . The virus strain , a 2001 human isolate from Iquitos , Peru , was obtained from the World Reference Center for Emerging Viruses and Arboviruses at the University of Texas Medical Branch . It was passaged once on Vero cells before RNA extraction . Details on primers and restriction sites are available upon request . To produce an attenuated MAYV that was capable of replicating in vertebrate cells , but not in invertebrate cells , the translation of viral structural proteins was placed under control of the EMCV IRES , directly downstream from the subgenomic promoter . The subgenomic promoter was also inactivated with 14 synonymous mutations using standard PCR-based mutagenesis methods ( Fig . 1 ) . These mutations were chosen to inactivate the promoter while preserving the amino acid sequence of the nsP4 C-terminus . A single PCR-derived amplicon containing mutated subgenomic promoter and IRES sequence was cloned into wt MAYV plasmid at SanDI – NcoI sites . The complete cDNA clone was sequenced to ensure that no errors occurred during PCR amplifications or cloning . Plasmid DNA was linearized with PacI prior to in vitro transcription , semi-quantified by gel electrophoresis , and recombinant viral RNA was electroporated into Vero cells using conditions described previously [20] . Titers of rescued wt MAYV and MAYV/IRES were both 4 . 0×107 PFU/mL at 28 h post electroporation . Cell culture supernatants were harvested 28 h post electroporation , centrifuged to pellet cell debris , and stored at −80°C . All mice were purchased from Charles River Laboratories ( Wilmington , MA ) . Animal studies were approved by the University of Texas Medical Branch Institutional Animal Care and Use Committee . African green monkey kidney ( Vero ) and human fetal lung fibroblast cells ( MRC-5 ) cells were purchased from the American Type Culture Collection ( ATCC , Manassas , VA ) and maintained in culture with Dulbecco's Modified Eagle's Medium ( DMEM ) supplemented with 5% fetal bovine serum ( FBS ) and gentamicin sulfate and incubated at 37°C with 5% CO2 . Aedes albopictus-derived C6/36 cells were maintained in DMEM supplemented with 10% FBS , 1% tryptose phosphate broth ( TPB ) solution , and an antibiotic mixture of penicillin/streptomycin at 29°C and 5% CO2 . Vero and MRC-5 cells were used to assess the replication kinetics of the MAYV/IRES vaccine candidate and wt MAYV . Cells were grown to 95% confluency in 6-well plates . Virus was added to each well at a multiplicity of infection ( MOI ) of 0 . 1 plaque forming units ( PFU ) /cell in triplicate and incubated with the cells for 1 h . The cells were then washed twice with phosphate buffered saline ( PBS ) to remove residual virus , and 2 mL of medium were added to each well . At designated timepoints ( 6 , 12 , 24 , 36 and 48 hours post infection ( hpi ) for Vero cells , and 24 , 48 , 72 , and 96 hpi for MRC-5 cells ) , the culture supernatant was harvested for virus titration by plaque assay , then fresh medium ( 2 mL ) was added to replace the volume . To assess the stability of the MAYV/IRES vaccine candidate , 5 passages were performed in duplicate on both Vero and C6/36 A . albopictus cells in T25 flasks , with the cells at 95% confluency before infection at a MOI of 0 . 1 PFU/cell . As a control , wt MAYV was also passaged . Vero cells were incubated at 37°C and 5% CO2 for 48 h , while the C6/36 Ae . albopictus cells were incubated at 29°C and 5% CO2 for 72 h . Culture supernatants were then collected and used to infect a new flask at the same MOI . Virus titers from each passage were measured by plaque assay . To evaluate the genetic stability of the MAYV/IRES vaccine candidate , viral genomes from Vero passages 3 and 5 of both MAYV/IRES and wt MAYV were fully sequenced . Viral RNA was extracted using a QIAamp Viral RNA Mini Kit ( Qiagen , Valencia , CA ) . This was followed by RT-PCR which was performed in a two-step reaction process involving SuperScript III One-Step RT-PCR System ( Invitrogen , Grand Island , NY ) in conjunction with Phusion High-Fidelity DNA Polymerase ( New England Biolabs , Ipswich , MA ) . PCR amplicon sizes were confirmed by gel electrophoresis and then purified by a QIAquick PCR Purification Kit ( Qiagen ) . A BigDye kit ( Applied Biosystems , Foster City , CA ) was then used to prepare the samples for Sanger sequencing . Thirty-nine overlapping amplicons were used to cover the entire genome; primer sequences are available from the authors . Infant outbred CD1 mice have been shown to develop disease similar to humans for the arthralgic alphavirus CHIKV [26] , and were therefore chosen as a model to evaluate the MAYV/IRES attenuation . Cohorts of six-day-old outbred CD1 mice were infected over the dorsum subcutaneously ( SC ) with 104 PFU , a dose used previously [26] , and were subsequently monitored daily for 10 days for survival and body weight . To evaluate immunogenicity , cohorts of adult 28-day-old CD1 mice were also infected SC with 105 PFU , and survival and body weights were monitored daily until day 28 post infection . Mice were bled on days 1–3 after infection , and serum was tested for viremia by plaque assay [27] to assess attenuation . On day 28 post infection , the animals were bled and a plaque reduction neutralization test ( PRNT ) was performed on the sera to measure antibodies as described previously [27] . MAYV produces no detectable disease in adult , immunocompetent mice . Therefore , to assess attenuation , cohorts of ca . 5–8-week-old interferon type I receptor-deficient A129 mice were infected intradermally ( ID ) on the left footpad ( FP ) with 104 PFU . The animals were monitored for survival , body weight changes , and viremia . Footpad swelling was also measured using a caliper at the site of inoculation . At day 28 post infection , sera were collected and PRNTs were performed . On day 29 post infection , the mice were challenged SC with 104 PFU of wt MAYV strain CH . The mice were monitored the following 7 days for survival , change in body weight , and viremia . The University of Texas Medical Branch ( UTMB ) Institutional Animal Care and Use Committee approved the animal experiments described in this paper under protocol 02-09-068 . UTMB complies with all applicable regulatory provisions of the U . S . Department of Agriculture ( USDA ) - Animal Welfare Act; the National Institutes of Health ( NIH ) , Office of Laboratory Animal Welfare - Public Health Service ( PHS ) Policy on Humane Care and Use of Laboratory Animals; the U . S Government Principles for the Utilization and Care of Vertebrate Animals Used in Research , Teaching , and Testing developed by the Interagency Research Animal Committee ( IRAC ) , and other federal statutes and state regulations relating to animal research . The animal care and use program at UTMB conducts reviews involving animals in accordance with the Guide for the Care and Use of Laboratory Animals ( 2011 ) published by the National Research Council . Analysis of variance ( ANOVA ) followed by a Tukey's post-hoc test , Kruskall-Wallis with Bonferroni correction for multiple comparisons , Kaplan-Meier , and Mann-Whitney test were performed using Prism 5 ( GraphPad Software , La Jolla , CA ) . P-values<0 . 05 were considered significant .
To assess the replication kinetics , virus derived from electroporated Vero cells was compared to wt MAYV after infection of Vero cells ( Fig . 2A ) . Infections were performed in triplicate ( n = 3 ) at a MOI of 0 . 1 PFU/cell . Both MAYV/IRES and wt MAYV titers peaked 36 hpi , but wt MAYV had a slightly higher titer of 1 . 1×108 PFU/mL while MAYV/IRES had a peak titer of 7 . 8×107 PFU/mL . Significant differences were seen only at the 48 hpi timepoint ( ANOVA , p<0 . 05 ) . Plaque morphology was consistent throughout the experiment , with wt MAYV having a slightly larger ( 0 . 5–3 mm ) and more diffuse plaque morphology than MAYV/IRES ( 0 . 5–2 mm ) under 0 . 4% agarose in MEM ( 48 h incubation ) . MRC-5 cells are well characterized and widely used in cell culture-based vaccine production . Therefore , we also measured the replication kinetics of the MAYV/IRES vaccine candidate , as well as wt MAYV on this cell line in triplicate wells ( n = 3 ) at a MOI of 0 . 1 PFU/cell ( Fig . 2B ) . The MAYV/IRES virus reached a peak titer of 106 . 7 PFU/ml at 72 hpi , which was much later and at a lower titer than wt MAYV . Plaque morphology measured on Vero cells of MAYV/IRES virus derived from MRC-5 or Vero cells was comparable . The stability of MAYV/IRES was tested in vitro by 5 serial passages in Vero cells , in duplicate at an MOI of 0 . 1 PFU/cell . MAYV/IRES maintained a slightly lower titer than wt MAYV throughout the passages , with a range of 4 . 2×107 PFU/mL after passage 2 , to a peak of 1 . 9×108 PFU/mL after passage 3; wt MAYV titers remained between 108 and 109 PFU/mL ( data not shown ) . To evaluate the genetic stability of the MAYV/IRES vaccine candidate , the complete consensus sequences of passages 3 and 5 were determined using overlapping amplicons generated by RT-PCR , and no mutations were detected . MAYV/IRES was also serially , blind passaged 5 times in C6/36 A . albopictus mosquito cells to confirm its lack of mosquito host range . As expected , the virus was not detected during any passage , while wt MAYV replicated to high titers ( data not shown ) . Cohorts of 6-day-old CD1 mice were infected SC with 104 PFU of either MAYV/IRES ( n = 14 ) , wt MAYV ( n = 15 ) , or were sham-infected with PBS ( n = 15 ) . Mice infected with wt MAYV began to die starting 3 dpi and complete mortality was seen by day 8 ( data not shown ) . All MAYV/IRES- and sham-infected mice survived until the study was terminated 10 days after inoculation . As expected , the wt MAYV-infected cohort did not gain weight as quickly as the MAYV/IRES- or sham-infected animals , and the average weight of wt-infected animals declined rapidly beginning 4 days post-infection . There was no significant difference in weight change between MAYV/IRES- and sham-infected animals ( Kruskall-Wallis with Bonferroni correction for multiple comparisons ) . Due to the high mortality in infant CD1 mice infected with wt MAYV , adult CD1 mice ( 28 days-old ) were also tested as a potential virulence model . Mice were infected SC with 105 PFU of either MAYV/IRES ( n = 10 ) or wt MAYV ( n = 10 ) , and negative controls were sham ( PBS ) -infected ( n = 6 ) . Unlike the infant 6-day-old CD1 mice , the 28-day-old mice all survived infection with wt MAYV until the study was terminated 28 days after infection . To assess with greater sensitivity signs of disease , the animals were weighed post-vaccination ( Fig . 3A ) . The MAYV/IRES- and sham-infected cohorts gained weight steadily throughout the experiment , while the wt MAYV-infected mice lost some weight initially , but recovered by day 5 post-infection , then proceeded to gain weight in a manner similar to the other cohorts . However , these differences in weight change were not significant ( p≥0 . 07 , Kruskall-Wallis with Bonferroni correction for multiple comparisons ) . To quantify viral loads of the MAYV/IRES vaccine candidate , viremia was assessed post-vaccination ( Fig . 3B ) . Both MAYV/IRES and wt MAYV produced a peak viremia titer at day 2 post-infection , but MAYV/IRES viremia was of shorter duration and of significantly lower mean peak titer , just over 103 PFU/mL , compared to 107 PFU/mL for wt MAYV . Serum neutralizing antibody titers were measured at 28 days post-infection using an 80% PRNT . MAYV/IRES titers ranged from 160 to ≥640 ( mean = ≥304 ) , and were not significantly different from those of wt MAYV-infected animals ( Kruskall-Wallis with Bonferroni correction for multiple comparisons ) ( Fig . 3C ) . A129 mice lack functional type 1 interferon receptors and are therefore a very sensitive model for human arthritic alphavirus infection [28] . They have been used as a lethal model for alphavirus vaccine safety and challenge studies [20] . Cohorts of adult A129 mice ( n = 8 ) were infected with MAYV/IRES or wt MAYV , or sham-infected with PBS . Injections were performed intradermally on the left footpad with 104 PFU . All MAYV/IRES- and sham-infected mice survived until the experiment was terminated on day 28 , while all wt MAYV-infected mice died by day 5 ( Fig . 4A ) . Both the MAYV/IRES and wt MAYV cohorts lost weight initially , but wt MAYV-induced loss was more dramatic and significantly greater than that of the MAYV/IRES-infected animals ( Fig . 4B ) ( p<0 . 01 , Kruskall-Wallis with Bonferroni correction for multiple comparisons ) . There was no significant difference in footpad swelling among cohorts until 3 days after infection , when wt MAYV-infected mice showed a large increase in footpad diameter , which was significantly greater than mean swelling of both MAYV/IRES- and sham-infected cohorts ( Fig . 4C ) ( p<0 . 01 , Kruskall-Wallis with Bonferroni correction for multiple comparisons ) . Viremia was measured post-vaccination to quantify the viral load ( Fig . 4D ) . Both MAYV/IRES and wt MAYV cohorts reached high titers in the peripheral blood , with MAYV/IRES peaking at day 3 post-infection with a titer of 5 . 5×108 PFU/mL and wt MAYV reaching a slightly higher titer of 1 . 4×109 PFU/mL . Differences were significant only on day one post-infection ( p<0 . 001 , Kruskall-Wallis with Bonferroni correction for multiple comparisons ) . At day 28 post-infection , 7 of the 8 MAYV/IRES-vaccinated A129 mice had neutralizing antibody titers ≥640 , while the remaining mouse had a titer of 320 ( mean = ≥604 ) . The mean PRNT antibody titer for A129 mice was significantly higher than that for CD1 immunocompetent mice ( Student's T-test , p<0 . 01 ) , possibly reflecting greater vaccine replication in the former ( although the ages were not exactly matched ) . The sham-vaccinated A129 mice ( n = 3 ) did not have detectable antibodies ( <20 ) . Mice were then challenged SC with 104 PFU of wt MAYV to assess the efficacy of the MAYV/IRES vaccine . All vaccinated mice survived , while all of the sham-vaccinated mice were dead by day 7 , representing a significant difference in mortality ( p<0 . 01 , Kaplan-Meier; see Fig . 5A ) . To monitor disease in a more sensitive manner , weight was measured post-vaccination ( Fig . 5B ) . The sham-vaccinated , challenged cohort lost weight more quickly and dramatically than the MAYV/IRES-vaccinated group ( p<0 . 01 , Mann-Whitney ) . To assess viral load , viremia post-challenge was also measured ( Fig . 5C ) . The MAYV/IRES-vaccinated group showed a decreased viremic response upon challenge compared to the sham-vaccinated animals , only reaching a mean titer of 2 . 0×102 PFU/mL at day 3 post-challenge , while the control group reached a much higher titer of 4 . 8×108 PFU/mL 3 days post-challenge ( p<0 . 05 , Mann-Whitney ) .
It has been over 60 years since the discovery of MAYV in Trinidad , and there is still no licensed vaccine available despite continued outbreaks , and the potential for urban transmission in a dengue-like cycle [5] , [12] that could expose millions of people . Our MAYV/IRES vaccine was designed to offer single-dose , rapid protection to protect people both in endemic regions and in the event of an urban outbreak . Previous attempts to generate a vaccine to protect against MAY focused on inactivated wt virus [18] . A single vaccination proved immunogenic in adult CD1 mice , and efficacy was demonstrated indirectly via passive transfer of the immune mouse sera to infant mice , followed by lethal challenge . However , no further testing of this vaccine has been reported . To capitalize on the advantages of live-attenuated vaccines , including rapid and long-lasting immunity as well as ease of manufacture , we used the IRES-based attenuation approach that has been demonstrated to offer highly stable and predictable attenuation for alphaviruses [19]–[24] . Unlike traditional alphavirus attenuation derived from cell culture passages that typically relies on unstable point mutations , resulting in reactogenicity and the potential for reversion to wt virulence and transmissibility [29]–[32] the IRES-based rationale approach suppresses structural viral protein expression by elimination of the subgenomic promoter using multiple inactivating mutations . Thus , reversion is highly unlikely because the promoter sequence is very specific and intolerant of change [33] , resulting in superior attenuation stability following serial mouse passages compared to traditional point mutation-dependent attenuation [22] . Further safety is achieved through the use of the encephalomyocarditis virus IRES , which inefficiently mediates translation in insect cells [25] , and thus eliminates the possibility for mosquito transmission . Finally , the titers of nearly 108 PFU/cell of MAYV/IRES produced by vaccine substrate-approved Vero cells should be adequate for large-scale manufacture , and the stability we demonstrated following Vero cell passages will be critical for licensure . Like previous studies using the IRES-based alphavirus attenuation approach , our results showed that MAYV/IRES is stable in cells of mammalian origin ( Vero ) , but incapable of efficient replication in a C6/36 A . albopictus cell line . Previous studies have showed that other IRES-based attenuated alphaviruses are also incapable of replication after intrathoracic inoculation into A . albopictus mosquitos [20] , [22] . In every murine model we tested , MAYV/IRES was highly attenuated , only producing minimal signs of disease in the highly stringent A129 model that cannot mount an effective interferon response . This vaccine candidate was also highly immunogenic , inducing high levels of neutralizing antibody titers in both adult CD1 and A129 mice at 28 days post-vaccination . Challenge of A129 vaccinated mice at 29 days post-infection with a high dose of wt MAYV showed complete protection from detectable disease , despite the high virulence and complete lethality of MAYV in unvaccinated animals . These murine studies indicate that MAYV/IRES is highly attenuated , highly immunogenic , and provides strong protection against MAYV challenge . Further studies in another animal model are needed . Typically , nonhuman primates such as macaques reproduce human-like disease after alphavirus infection [24] , [34]–[39] . These animals should be evaluated as models for human MAYV to determine if they will be useful for the next steps in preclinical evaluation of MAYV/IRES . A variety of alternative vaccine development approaches are available for alphaviruses including inactivated virus , subunit protein , DNA and virus-like particles ( VLP ) as well as traditionally attenuated and chimeric vaccines [40] , [41] . All of these approaches emphasize safety but have significant drawbacks including a multiple dose requirement for efficacy , short-lived immunogenicity necessitating boosters , challenging delivery ( DNA via electroporation ) and complex , expensive manufacture ( VLPs ) and the risk of residual live virus after inactivation , which was shown to result in the death of an eastern equine encephalitis-vaccinated horse in California [42] . Our MAYV/IRES candidate overcomes all of these shortcomings to generate rapid immunity following a single dose , and should have greatly reduced reactogenicity due to its robust , highly stable attenuation design . Although further testing should be done to evaluate the duration of protective immunity , other IRES-based alphavirus vaccines have generated completely protective immunity in macaques for over one year ( C . Roy , S . C . W . , unpublished ) . MAYV/IRES therefore should be ideal for inducing rapid , long-lived immunity after a single dose for use in developing countries where MAYV is endemic , as well as for a traveler's vaccine for persons visiting South America . In summary , our MAYV/IRES vaccine candidate is highly attenuated and immunogenic , unable to infect mosquito cells , and provides protection from lethal challenge in murine models . These results indicate that further preclinical development of MAYV/IRES is justified for its evaluation as a potential human vaccine that could protect people from MAY in South America , but also on other locations if the virus spreads and urbanizes like the closely related CHIKV [5] , [43]–[46] . Furthermore , MAYV/IRES should be evaluated for its ability to protect against CHIKV and Ross River viruses , other closely related alphaviruses that cause epidemics in Africa and Asia , or Australia and Oceania , respectively . CHIKV is of particular concern because in December of 2013 it invaded the Caribbean , representing the first autochthonous transmission in the Western Hemisphere [47]–[49] . This event could portend a major epidemic throughout the Americas if spread to the mainland occurs into dengue-endemic regions where both A . aegypti and A . albopictus mosquito vectors are present along with a nearly naïve human population . The latter vector is highly susceptible to Asian CHIKV strains with adaptive mutations that dramatically enhance its vectorial capacity [50]–[55] , and it is unknown if similar mutations could enhance MAYV urbanization in a similar manner . An effective vaccine could greatly mitigate these risks and have a major impact on public health in South America .
|
Mayaro virus ( MAYV ) is a mosquito-borne alphavirus that causes severe and sometimes chronic arthralgia in persons in South America , where it circulates in forest habitats . It is widely neglected because it is typically mistaken for dengue due to the overlap in the clinical signs and symptoms , and the lack of laboratory diagnostics in most endemic locations . Furthermore , MAYV has the potential to initiate an urban transmission cycle like that of dengue , which could result in a dramatic increase in human exposure . Because there is no effective vaccine or specific treatment , we developed a candidate vaccine to protect against MAYV infection . We used an attenuation approach based on the elimination of the MAYV subgenomic promoter and insertion of a picornavirus internal ribosome entry site to mediate translation of the structural proteins . This vaccine was well attenuated in mouse models , highly immunogenic , and protected against fatal MAYV infection . Our results indicate that this MAYV strain is promising for further development as a potential human vaccine .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"and",
"occupational",
"health",
"medicine",
"and",
"health",
"sciences",
"global",
"health",
"biology",
"and",
"life",
"sciences",
"microbiology"
] |
2014
|
A Novel Live-Attenuated Vaccine Candidate for Mayaro Fever
|
New strategies for collecting post-mortem tissue are necessary , particularly in areas with emerging infections . Minimally invasive autopsy ( MIA ) has been proposed as an alternative to conventional autopsy ( CA ) , with promising results . Previous studies using MIA addressed the cause of death in adults and children in developing countries . However , none of these studies was conducted in areas with an undergoing infectious disease epidemic . We have recently experienced an epidemic of yellow fever ( YF ) in Brazil . Aiming to provide new information on low-cost post-mortem techniques that could be applied in regions at risk for infectious outbreaks , we tested the efficacy of ultrasound-guided MIA ( MIA-US ) in the diagnosis of patients who died during the epidemic . In this observational study , we performed MIA-US in 20 patients with suspected or confirmed YF and compared the results with those obtained in subsequent CAs . Ultrasound-guided biopsies were used for tissue sampling of liver , kidneys , lungs , spleen , and heart . Liver samples from MIA-US and CA were submitted for RT-PCR and immunohistochemistry for detection of YF virus antigen . Of the 20 patients , 17 had YF diagnosis confirmed after autopsy by histopathological and molecular analysis . There was 100% agreement between MIA-US and CA in determining the cause of death ( panlobular hepatitis with hepatic failure ) and main disease ( yellow fever ) . Further , MIA-US obtained samples with good quality for molecular studies and for the assessment of the systemic involvement of the disease . Main extrahepatic findings were pulmonary hemorrhage , pneumonia , acute tubular necrosis , and glomerulonephritis . One patient was a 24-year-old , 27-week pregnant woman; MIA-US assessed the placenta and provided adequate placental tissue for analysis . MIA-US is a reliable tool for rapid post-mortem diagnosis of yellow fever and can be used as an alternative to conventional autopsy in regions at risk for hemorrhagic fever outbreaks with limited resources to perform complete diagnostic autopsy .
In the last decades , we have been globally identifying an unprecedented number of emerging infections [1] . As a result , considerable effort is currently being made by several actors from governmental and non-governmental institutions for better preparedness for the next expected emerging threats to public health worldwide [2–4] . Therefore , there is great interest in the development of tools for the early detection of infectious outbreaks for the establishment of appropriate measures [2 , 5] . Autopsy has been an important and indisputable diagnostic tool for the detection of novel diseases . In the past years , our group has performed autopsy studies to describe new aspects of human pathology of emerging infectious diseases , such as measles , leptospirosis , and influenza A ( H1N1 ) pdm09 [6–8] . However , the global distribution of facilities capable of performing the procedure is markedly uneven . New strategies for collecting post-mortem tissue samples are necessary , particularly in areas where outbreaks of infectious diseases are occurring and where the identification of the causative agent as well as its effects on target organs is a public health priority [4 , 5 , 9] . Minimally invasive autopsy ( MIA ) has been proposed as an alternative to conventional autopsy , conceived as targeting small diagnostic biopsies by needle puncture of key organs , with or without the guidance of any imaging technique . The use of computed tomography ( CT ) and CT-angiography , associated with needle biopsy , has been proposed as feasible for diagnosis of common causes of death [10 , 11] . However , these techniques can only be performed in centers of excellence , requiring substantial budgets . More recently , ultrasound-guided tissue sampling ( MIA-US ) has also been tested , with promising results [12–18] . This methodology represents a portable , rapid and low-cost post-mortem technique , which may be especially useful in countries where mortality data are largely unavailable [2 , 13–18] . Viral hemorrhagic fevers ( VHFs ) are severe viral infections that may present as hemorrhagic disease with fatal multi-organ failure . In endemic areas , they can cause long-lasting epidemics with great impact on human morbidity and mortality . Yellow fever ( YF ) , in particular , is a re-emerging disease , endemic in tropical regions of South America and sub-Saharan Africa . It is a mosquito-borne flavivirus-induced VHF , with a high case-fatality rate , clinically manifested as hepatic dysfunction , renal failure , coagulopathy , and shock [19] . The investigation of deaths related to VHFs should put MIA-US into perspective among other potential diagnostic strategies . From the end of 2017 to mid-May 2018 , we experienced a YF epidemic in the southern region of Brazil . During this period , 498 autochthonous confirmed YF cases were registered in the state of Sao Paulo , with 176 deaths ( fatality rate: 35 . 4% ) . A substantial part of the fatalities ( 80 cases ) was referred to the autopsy service at Sao Paulo University Medical School . Aiming to provide new information on low-cost post-mortem techniques that could be applied in at-risk regions in Brazil as well as worldwide , we tested the efficacy of MIA-US in the post-mortem diagnosis of 20 patients who died with suspected or confirmed YF during the recent epidemic in Brazil .
This prospective observational study was approved by the University of Sao Paulo School of Medicine Internal Review Board ( CAAE protocol number: 18781813 . 2 . 0000 . 0068 ) . Informed written consent was obtained from the next of kin . From January 23 , 2018 to February 27 , 2018 , 20 deceased patients with suspected or confirmed YF underwent MIA using ultrasound-guided percutaneous core needle biopsy . Conventional autopsy ( CA ) was performed subsequently by a distinct group of pathologists , blinded to the MIA-US results . Fig 1 illustrates the sequence of the autopsy procedures . All the patients included in the study fulfilled the following criteria: 1 ) patients died with suspected or confirmed yellow fever; 2 ) an autopsy was requested by the clinician; 3 ) death occurred during business hours , when interventional radiologists were available to perform the MIA-US; 4 ) informed consent to perform the autopsy was given by a family member . The case definition of YF employed during the epidemic period was established by the Brazilian Ministry of Health and the Health Department of the State of Sao Paulo . Suspected cases referred to those patients who had a sudden onset of high fever associated with jaundice and/or hemorrhages , who lived or had visited areas with cases of YF , YF epizootics in non-human primates , or isolation of yellow fever virus ( YFV ) in vectors , regardless of the vaccine status for YF , during the preceding 15 days . Confirmed cases referred to those patients who had compatible clinical presentation and laboratory confirmation by at least one of the following methods: positive serum IgM ( MAC-ELISA ) ( performed in 10 of 20 patients ) ; detection of YFV-RNA by RT-PCR in blood samples ( performed in 17 of 20 patients ) ; and histopathology compatible for YF hepatitis with detectable YF antigen in tissues by immunohistochemistry ( wild , SP strain , hyperimmune , IAL–SP , Brazil ) [20] . In all of the cases , viral hepatitis ( A , B , and C ) and dengue fever were excluded by serology and/or RT-PCR . The procedures were performed at the “Image Platform in the Autopsy Room” , a research center in the University of Sao Paulo Medical School , located next to the Autopsy Service of Sao Paulo University ( https://pisa . hc . fm . usp . br/ ) . Radiologists and pathologists were aware of the information provided by the next of kin prior to autopsy . All US examinations were performed by interventional radiology physicians , who were trained to perform gray-scale post-mortem US during a period of 8 weeks . In total , fifteen exams were necessary in order to achieve thorough ultrasound imaging and good quality biopsy samples . We used a SonoSite M-Turbo portable ultrasound ( Fujifilm , Bothell , WA , USA ) with broadband and multifrequency transducers: C60x ( 5–2 MHz Curved ) and HFL38X ( 13–6 MHz Linear ) and DICOM medical images . Tissue samples from the liver ( at least two samples ) , lungs ( three samples from each lung ) , both kidneys ( one sample each ) , spleen ( one sample ) , and heart ( one sample ) were collected under ultrasound guidance using a large core ( 14-gauge ) needle biopsy . A final MIA-US diagnosis was provided by conjunct analysis of US and biopsy data , blinded to CA results . The autopsies were performed following the Letulle technique , where all the organs are removed en masse , requiring dissection of each organ [21] . Tissue samples were collected from all body systems . Both biopsy and autopsy samples were submitted for routine histological examination , and stained with hematoxylin-eosin ( H&E ) . Brown-Brenn , Ziehl-Neelsen , and Grocott stains were used to detect bacteria , acid-fast bacilli , and fungi , respectively , when required . Eleven liver biopsy samples and all autopsy liver samples were submitted for reverse transcriptase-polymerase chain reaction ( RT-PCR ) for detection of YFV-RNA . Liver samples measuring 1 cm3 were stored at −70°C . The tissue was macerated , the nucleic acid extraction was performed using the TRIzol reagent ( Life Technologies , Carlsbad , CA , USA ) , and carried out according to the manufacturer’s instructions . Molecular detection of YFV was performed using the AgPath-ID One-Step RT-PCR Reagents ( Ambion , Austin , TX , USA ) with specific primers/probe previously described [22] . To identify cases of adverse vaccine response ( i . e . , fatal cases associated with the vaccine virus ) we used primers/probe specific for the vaccine virus [23] . qRT-PCR reactions consisted of a step of reverse transcription at 45°C for 10 min , enzyme activation at 95°C for 10 min , and 40 cycles at 95°C for 15 s and 60°C for 45 s for hybridization and extension with the use of ABI7500 equipment ( Thermo Fisher Scientific , Waltham , MA , USA ) . All biopsy and autopsy liver samples were submitted for IHC for detection of YFV antigens as previously described [24] . Briefly , after antigen retrieval using Tris-EDTA at pH 9 . 0 , the sections were incubated overnight with the primary antibody ( goat anti-human IgM polyclonal anti-arbovirus , provided by Institut Pasteur de Dakar , Dakar , Senegal , 1:20 , 000 ) at 4°C [25] . The slides were incubated with biotinylated secondary antibody ( Reveal–Biotin-Free Polyvalent HRP DAB , Spring Bioscience , cod . SPD-125 ) and chromogen ( Dako Liquid DAB+ Substrate Cromogen System , Dako , cod . K3460 ) and counterstained with Harris-hematoxylin ( Merck , Darmstadt , Germany ) . The primary antibody was tested in liver samples from patients with virus hepatitis ( A , B , and C ) , herpes virus , cytomegalovirus , adenovirus , dengue virus , Treponema pallidum , Leptospira , and atypical mycobacteria infections , obtained from our autopsy archive . The IHC reaction resulted negative in all these samples . We compared the MIA-US diagnosis with the CA diagnosis of the cause of death and main disease . In addition , we performed a descriptive analysis of the concordances and discrepancies between the diagnoses of MIA-US and CA in all organs analyzed .
The patients comprised 13 men and 4 women , with a median age of 47 ( 24–86 ) years . Fourteen patients died at our institution and three were referred for autopsy . Sixteen patients lived or traveled to the epizootic areas of the city . One patient lived in urban area and had YF vaccine-associated viscerotropic disease ( YEL-AVD ) , which was confirmed by liver RT-PCR . Six patients received YF vaccination 0–3 days before death; only one of them presented with YEL-AVD . One patient was a 24-year-old , 27-week pregnant woman . The median timespan between the occurrence of symptoms and hospital admission was 4 days ( 1–15 ) and that between the occurrence of symptoms and death was 8 days ( 6–21 ) . The most frequent associated clinical conditions were alcoholism ( n = 7 ) , smoking ( n = 5 ) , and systemic arterial hypertension ( n = 4 ) . One patient had undergone liver transplantation three days before death for hepatic failure due to YF . The primary symptoms upon hospital admission were fever , jaundice , myalgia , anorexia , abdominal pain and hemorrhagic phenomena . The main alterations in the laboratory data were related to acute liver failure , as well as renal dysfunction and metabolic acidosis . Table 1 presents detailed clinical data of the 17 patients . Adequate biopsy samples from the liver ( 100% ) , kidneys ( 94% ) , and lungs ( 88% ) were obtained in majority of patients . More limited samples were obtained from the spleen ( 82% ) and heart ( 76% ) . Liver biopsies showed panlobular hepatitis with severe steatosis and midzonal apoptotic bodies , which were characteristic of fatal YF , in all samples analyzed ( 100% ) . All liver samples were submitted for IHC and 11 liver samples were submitted for RT-PCR for detection of YFV . IHC revealed positive results in 16 samples ( 94% ) , and RT-PCR revealed positive results in all 11 samples ( 100% ) . Lung samples showed pulmonary involvement ( 73% ) that included alveolar hemorrhage ( 67% ) and pneumonia ( 60% ) . Pulmonary aspergillosis was confirmed in a single lung biopsy . Kidney alterations included acute tubular necrosis ( 100% ) and a mesangial proliferative glomerulonephritis ( 63% ) . Spleen biopsies showed lymphoid hypoplasia ( 93% ) , splenitis ( 86% ) , and hemophagocytosis ( 43% ) . Heart biopsies showed interstitial edema ( 85% ) and fiber hypertrophy ( 54% ) . In the pregnant woman , MIA-US could be used to assess the placenta and provided adequate placental tissue for histological analysis . The cause of death in the 17 patients was hepatic failure . Associated organ-specific hemorrhages and/or hemorrhagic shock was present in 16 patients . One patient presented with septic shock . All patients presented with panlobular hepatitis with severe steatosis and midzonal apoptotic bodies . IHC was positive in 15 ( 88% ) of 17 liver samples , and RT-PCR was positive in 17 ( 100% ) liver samples . One patient had undergone liver transplantation after developing YF-induced hepatic failure and the graft was infected . One patient presented with YEL-AVD , confirmed by liver and spleen RT-PCR . Other autopsy liver findings were hepatomegaly , ischemic centrilobular necrosis , and alcoholic cirrhosis ( two cases ) . Ascites was detected in 10 ( 59% ) patients . Table 2 presents detailed liver findings at MIA-US and CA for the 17 patients . At the respiratory system , alveolar hemorrhage ( 94% ) and pneumonia ( 53% ) were the main pulmonary autopsy findings . Pleural effusion ( 41% ) , diffuse alveolar damage ( 41% ) , and bronchoaspiration ( 18% ) were also observed . Two patients presented with fungal pneumonia ( one aspergillosis and one mucormycosis ) and associated pulmonary necrosis secondary to mycotic thrombus . Two patients presented with pulmonary embolism . The main kidney alterations were acute tubular necrosis ( 94% ) and mesangial proliferative glomerulonephritis ( 88% ) . Other kidney findings were interstitial fibrosis , nephrosclerosis , hypertensive nephropathy , vascular thrombosis , and pyelonephritis . Spleen findings included lymphoid hypoplasia ( 100% ) , splenitis ( 94% ) and cytophagocytosis ( 50% ) . One patient underwent splenectomy . The primary cardiac changes at autopsy included fiber hypertrophy ( 76% ) , interstitial edema ( 71% ) , myocardiosclerosis ( 65% ) , and coronary atherosclerosis ( 65% ) . Autopsy also showed organ alterations that were not assessed by MIA-US , such as pancreatic changes ( ischemic pancreatic changes , peripancreatic steatonecrosis , and alcoholic pancreatic interstitial fibrosis ) , central nervous system ( CNS ) changes ( cerebral edema , perivascular hemorrhage , and cerebral herniation ) , gastrointestinal ( GI ) bleeding , skin perivascular inflammation , lymphoid hypoplasia in lymph nodes , and bone marrow with erythroid depletion and hemophagocytosis . One patient was a 19-year-old man who died due to perforated appendicitis and sepsis . Purulent fluid leaked from the abdominal cavity when the radiologist performed US-guided liver and kidney biopsies , indicating an intra-abdominal infection . Lung biopsy in this case showed secondary acute lung injury with intense pulmonary hemorrhage . One patient was an 87-year-old woman with sepsis due to bacterial pneumonia and pyelonephritis , both observed respectively on lung and kidney US-guided biopsies . The third patient was a 14-year-old boy with acute myeloid leukemia . Liver and kidney samples from MIA-US showed infiltration of atypical myeloid CD34 positive cells ( blasts ) .
The present results demonstrate the effectiveness of MIA-US in the post-mortem diagnosis of the main disease ( YF ) and cause of death ( hepatic failure ) in a group of patients evaluated during the epidemic of YF that occurred in early February 2018 in Sao Paulo , Brazil . We found a 100% concordance between MIA-US and CA for the diagnosis of the main disease and cause of death , which validated our initial proposition that this diagnostic method could be a rapid and viable alternative to CA in determining post-mortem diagnosis of a VHS . The excellent correlation between MIA-US and CA in this group of patients may be explained by some factors . First , the acknowledged accuracy of ultrasound to guide needles into the viscera affected by this viral disease , especially the liver , kidneys , and lungs . Second , important technical aspects were applied to obtain tissue samples , involving the use of the portable ultrasound equipment and Tru-Cut 14G needles . Third , procedures were performed by interventional radiologists after a training period . The ultrasound guidance , ensuring the researchers about the organ or region that underwent needle biopsies , favored good quality samples in most patients , thus allowing precise histopathologic diagnosis to be made . Technical limitations were observed in the access to the spleen and heart of some patients , due to their anatomical locations , i . e . the heart being partially obscured by the sternum and the spleen by the rib cage and intestinal gas distension . The procedure turned out to be a simple and reliable method of post-mortem tissue sampling that could be applied in remote areas after adequate training . Our results are in concert with those of other researchers [12–18 , 26 , 27] . In the series by Fariña et al . [12] , there was an 83% diagnostic concordance between MIA-US and CA . Castilho et al . [14] reported a success rate of 83% for the determination of cause of death by MIA-US in a study of 30 patients with a predominance of infectious diseases . Recently , Martinez et al . [15] compared MIA-US and CA in 30 patients and obtained an 89 . 5% concordance for the detection of an infectious agent directly involved in the cause of death . Conversely , Cox et al . [13] found a 57% concordance between MIA-US and CA for infectious major diagnoses , mainly present in the lungs ( 63% ) , liver ( 44% ) , and spleen ( 34% ) . Result differences among studies possibly relate to different patient populations and different causal mixes . Our results further showed that MIA-US is also a reliable tool to obtain samples with sufficient quality for molecular studies and microorganism identification and to assess the systemic involvement of the disease . Besides typical hepatic changes observed on histological examination , IHC and/or RT-PCR allowed definite viral detection in 94% of the patients . These results encourage further development of local technologies in endemic regions , which might provide access to real-time research data , improving epidemic preparedness . In addition to the liver , the main affected organs at CA were the lungs and kidneys [19] . Lung biopsies showed that pulmonary hemorrhage and suppurative pneumonia were important complications of YF infection , significantly contributing to death . Besides the more prevalent bacterial pneumonia , MIA-US was able to detect severe pulmonary aspergillosis in one patient , demonstrating an immunosuppressive state associated with fatal YF . Shock-induced acute tubular necrosis was related to kidney dysfunction and could be detected in all kidney biopsies . Interestingly , in one patient with pyelonephritis , the etiologic agent ( Aspergillus spp . ) could be identified at MIA-US but not at CA . YF-induced glomerulonephritis could be observed in most kidney biopsies [24] . One of the 17 patients was a 24-year-old , 27-week pregnant woman . In this specific case , MIA-US was able to assess the placenta and provided adequate placental tissue for analysis . Therefore , we believe that MIA-US can be an important tool to study maternal deaths and maternal-fetal transmission . The limitations of MIA-US in the present study were related to changes observed at autopsy that were not assessed by MIA-US . These mainly included changes related to GI tract and CNS . Shock-induced ischemic pancreatic alterations , steatonecrosis , and gastrointestinal bleeding , as well as cerebral edema , were common and important findings that certainly contributed to death . Therefore , for diseases that primarily affect the CNS and/or GI tract , strategies to include assessment of these organs should be considered . In the past years , we have been globally identifying an unprecedented number of emerging infections [1 , 4 , 5 , 18] . This phenomenon is probably associated with a complex interaction of factors such as human behavior , environmental changes and vector proliferation worldwide [1] . Precise post-mortem diagnosis during the first periods of an emerging epidemic would represent an improvement in identifying the specific etiologic agent , with substantial impact in disease monitoring . For instance , it is quite plausible that if post-mortem brain sampling had been performed in the early cases of microcephaly in Northeastern Brazil , the identification of Zika virus infection could have been made earlier [28] . Moreover , protective measures could have been taken more effectively . However , such measures are hindered by the lack of autopsies in areas prone to be affected by emerging infectious agents . In addition , because of its minimal invasiveness , MIA-US may represent a safer procedure to health authorities who investigate the emerging infectious diseases of high ( Ebola , for instance ) or unknown lethality . In this scenario , our results indicate that MIA-US represents a portable , low-cost post-mortem technique that can be useful for the rapid monitoring and surveillance of infectious diseases spread and outbreaks . The adoption of consolidate protocols , associated with international partnerships to use the resources of molecular diagnosis to regions where such resources are scarce , may , theoretically , expand the horizons of surveillance of global infectious threatens . In conclusion , our results show that MIA-US is a reliable tool for the post-mortem diagnosis of YF and can be used as an alternative to conventional autopsy in regions at risk for viral hemorrhagic fever of different etiologies .
|
Reliable mortality information is of paramount importance to establish sound public health policies . Autopsy is an important tool not only for determining the cause of death , but also for the detection of novel diseases . In the last decades , we have been globally identifying an unprecedented number of emerging infections . Therefore , there is great interest in the development of less invasive and low-cost tools for the accurate post-mortem diagnosis in fatal cases . Minimally invasive autopsy ( MIA ) , conceived as targeting diagnostic biopsies of key organs by needle puncture , has been proposed as an alternative to conventional autopsy ( CA ) for the determination of cause of death in developing countries . In this research , we tested the efficacy of MIA in the post-mortem diagnosis of 20 patients with suspected or confirmed yellow fever who died during the recent epidemic of yellow fever that occurred in Brazil . There was a perfect agreement between MIA and CA in determining the cause of death ( hepatic failure ) and main disease ( yellow fever ) in all patients with confirmed yellow fever . This finding indicates that MIA can be used as an alternative to CA in regions at risk for infectious disease outbreaks with limited resources to perform conventional autopsies .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"reverse",
"transcriptase-polymerase",
"chain",
"reaction",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"biopsy",
"spleen",
"liver",
"diseases",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"autopsy",
"signs",
"and",
"symptoms",
"gastroenterology",
"and",
"hepatology",
"molecular",
"biology",
"techniques",
"kidneys",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"artificial",
"gene",
"amplification",
"and",
"extension",
"molecular",
"biology",
"yellow",
"fever",
"diagnostic",
"medicine",
"anatomy",
"physiology",
"hemorrhage",
"biology",
"and",
"life",
"sciences",
"renal",
"system",
"viral",
"diseases",
"fatty",
"liver",
"vascular",
"medicine",
"polymerase",
"chain",
"reaction"
] |
2019
|
Ultrasound-guided minimally invasive autopsy as a tool for rapid post-mortem diagnosis in the 2018 Sao Paulo yellow fever epidemic: Correlation with conventional autopsy
|
Shaping the primordia during development relies on forces and mechanisms able to control cell segregation . In the imaginal discs of Drosophila the cellular populations that will give rise to the dorsal and ventral parts on the wing blade are segregated and do not intermingle . A cellular population that becomes specified by the boundary of the dorsal and ventral cellular domains , the so-called organizer , controls this process . In this paper we study the dynamics and stability of the dorsal-ventral organizer of the wing imaginal disc of Drosophila as cell proliferation advances . Our approach is based on a vertex model to perform in silico experiments that are fully dynamical and take into account the available experimental data such as: cell packing properties , orientation of the cellular divisions , response upon membrane ablation , and robustness to mechanical perturbations induced by fast growing clones . Our results shed light on the complex interplay between the cytoskeleton mechanics , the cell cycle , the cell growth , and the cellular interactions in order to shape the dorsal-ventral organizer as a robust source of positional information and a lineage controller . Specifically , we elucidate the necessary and sufficient ingredients that enforce its functionality: distinctive mechanical properties , including increased tension , longer cell cycle duration , and a cleavage criterion that satisfies the Hertwig rule . Our results provide novel insights into the developmental mechanisms that drive the dynamics of the DV organizer and set a definition of the so-called Notch fence model in quantitative terms .
Patterning processes in multicellular organisms rely on faithful mechanisms of cell segregation and segmentation [1] , [2] . These ideas are beautifully illustrated by the morphogenetic events that the imaginal discs of Drosophila undergo during metamorphosis [3]–[7] . There , the combined action of heritable selector genes confers location identities at the single cell level [8] , [9] . For example , in the wing imaginal disc , engrailed and apterous genes endow cells with a posterior and a dorsal character respectively . Moreover , these genes grant some properties that determine cellular interactions that in turn restrict their locations , e . g . affinity and adhesion [2] , [10]–[12] . Thus , cells under control of selector genes cannot intermingle freely and their positions become restricted to regions within the primordium: the so-called compartments [10] , [11] , [13]–[15] . The concept of compartment implies the existence of non-trivial boundaries that control cell migration [16]–[19] . While these lineage frontiers are not necessarily associated with morphological hallmarks of the organism , they play in all cases an additional and all-important role for setting the developmental plan . Such task is first driven by the differential gene expression pattern at both sides of the compartments interface , i . e . selector gene activity on versus off , that induces short range signaling between cells and promotes further patterning [8] , [9] , [14] , [17] , [19] . In particular , a cellular population becomes specified by the boundary defining the so-called organizer . Subsequently , signaling by morphogens towards the compartments takes place [3] , [14] , [20] . As a result , cells at the compartment bulk “read” the generated morphogen concentration gradient and obtain positional information [3] , [21]–[25] . Therefore , an organizer acts in practice as the coordinate axis of a reference system . To this end an organizer must display some key features to guarantee its reliability as a source of positional information: the width of this cell population is constricted to few ( two , three ) cells [26] , [27] and they develop maintaining a straight shape [28] , [29] . Altogether , these findings meant a major breakthrough in modern Developmental Biology because of its powerful conceptual implications in terms of the modular design of multicellular organisms , conserved in both vertebrates and invertebrates , and its genetic foundation . Since the discovery of developmental compartments , almost forty years ago , much progress has been attained with regard to the processes that lead to their formation and function [8]–[10] , [26] , [28]–[32] . Our recent contributions include the reverse engineering of the gene regulatory network that is responsible for the robust and stable patterning of the dorsal-ventral ( DV ) organizer and the elucidation of its minimal underlying network motif , the so-called spatial toggle switch [27] , [33] . Thus , the underlying mechanisms that set the expression and activity pattern of the DV organizer and neighboring cells are now mostly clear from the point of view of a static tissue . Yet , many aspects of the functioning of the DV organizer of the wing imaginal disc still remain puzzling . In particular , how this pattern can be progressively and robustly scaled as cell proliferation advances remains a conundrum . Addressing such complex problem in an effective manner requires to transcend the molecular level and focus on its effects in terms of the mechanical interaction between cells . From that perspective , herein we propose an in silico framework that sheds light into the dynamics that shape the DV organizer for being an effective source of positional information and a cellular lineage controller . Based on the available experimental data , our approach falls into the field of Modeling and Computational Biology and introduces a realistic and novel description of the cellular dynamics of the DV organizer and neighboring compartments , leading to a series of quantitative predictions that can be experimentally tested . Within the aforementioned general developmental script about the formation and function of boundaries and organizers , some relevant peculiarities depend on the problem under consideration . Thus , in the case of the wing imaginal disc , the anterior-posterior ( AP ) organizer becomes established at the anterior side of the AP boundary , whereas the DV organizer is located at both sides of the boundary [17] , [34] , [35] ( see Fig . 1 ) . This has important implications with respect to the organizers characteristics and their role . In both cases the boundary is formed at the interface of two compartments . Still , the AP boundary strictly separates two cellular populations –the AP organizer , that belongs to the A compartment , and the P compartment– whereas the DV boundary develops embedded within the DV organizer population and the entity that keeps cells segregated from opposite compartments is the organizer itself rather than the boundary [17] , [34] , [35] . In other words , the separation between anterior and posterior cell populations is driven by an interface , namely the AP boundary , whereas the dorsal and vental cell populations are kept segregated by a cellular structure , namely the DV organizer . A fluid simile: in a two fluids mixture , the interface separating those would correspond to the AP boundary , whereas the DV organizer case would be comparable to a fluid film that intercalates between them . The mirror image nature of the DV organizer with respect to the compartments is due to a symmetric short signaling between compartments that sets the activation onset of the transmembrane receptor Notch [26] , [27] . It has been suggested that such symmetry is necessary from a morphological point of view since the D and V compartments lead , upon development , to the specular dorsal and ventral surfaces of the wing blade in the adult organism . On the other hand , the DV organizer and neighboring cells give rise to the wing margin [16] . Other relevant differences between the AP and DV cases refer to necessary and sufficient conditions for their establishment . Whereas in the AP case the differential cell affinity between A and P cells , driven by the engrailed selector gene , seems to be required for lineage restriction , in the DV case the gene expression signature of the organizer ( Notch activity ) is necessary and sufficient for establishing a lineage barrier regardless of the identity of the cell populations [28] , [29] , [36] , [37] . As a consequence , ectopic activation of Notch at either dorsal or ventral compartments recreates a functional organizer and , conversely , if Notch signal is blocked , then compartment cells can freely mix [36] . While there is an apparent contradiction in regards of the function of Notch for the maintenance of the DV organizer , transcriptional [38] versus non-transcriptional [28] –see also [31]– , it is clear that Notch receptor and its signaling pathway are indispensable elements for the establishment and maintenance of the DV organizer . Everything considered , researchers have adequately coined the term Notch fence model to describe these specific features [17] . Recent research has pointed out that mechanical effects play a central role in the function of the DV organizer . Thus , it has been shown that both F-actin and Myosin II accumulate by the zonula adherens at the junctions of the DV border [28] , [29] . Running along the boundary , these components putatively promote cell adhesion and increase the cortical tension of cells . In agreement with these studies , it has been recently reported that actomyosin-based barriers are effective inhibitors of cell mixing in other developmental stages of Drosophila [32] , [39] . These results provide evidence in favor of a crucial and active role of the cytoskeleton , and consequently of the mechanical effects , for keeping the straightness and fence-like features of the DV organizer . Importantly , it has been recently proved that , in addition to the differences in cell affinity , some of these contributions –increased cell tension– underlie the functioning of AP boundary too [39] . Moreover , other studies indicate that the consideration of dynamical and morphological factors related with the cell cycle is also required for understanding the stability and robustness of the DV organizer . During the course of development the increase in cell number of the wing imaginal disc is approximately 1000-fold . This poses the intriguing question of how the DV organizer deals with division events for maintaining its straightness , width , and stability . That is , how does the DV organizer pattern become robustly scaled as proliferation progresses ? Related to that , it has been demonstrated in different contexts that the orientation of cell divisions determines the shape of developing tissues and organs [40]–[45] . In particular , it is now clear , either from measurements of the orientation of the mitotic spindle or of the post-mitotic cellular allocation , that cells of the DV organizer follow a division pattern that is different from cells at the bulk of the compartments , favoring the division plane to be perpendicular to the DV boundary [28] , [41] . Nowadays it is widely recognized that in silico experiments are a powerful and effective tool for studying the dynamics of epithelial tissues like the wing imaginal disc [39] , [45]–[52] . Following the seminal study of Weliky and Oster [53] , the so-called vertex model was first introduced by Nagai and Honda [54] . The model exploits the polygonal-like morphology , the monolayer character , and the apicobasal mechanical polarization of epithelial cells to characterize them by a reduced set of points: the apical vertices . The dynamics of each cell vertex depends on the applied forces that derive from mechanical considerations , e . g . cytoskeleton activity . In the literature different examples are found where the vertex model has successfully described the wing imaginal disc . Recent advances include its packing , the AP compartmentalization [39] , [50] , the effects of the mechanical feedback on the tissue topology [52] , and the alignment of the planar cell polarity domains with the proximal-distal axis of the wing [45] . However , to the best of our knowledge , no example has been reported so far where a realistic dynamics of the cell cycle , the cell growth , and the division events are also taken into account . Within this framework our objectives are twofold . First , we propose an improved methodological approach for in silico experiments on epithelial tissue dynamics . To this aim , we present a simulation code based on the vertex model that includes the aforementioned dynamic and morphological effects in a realistic manner , i . e . cell cycle , growth , and cleavage criterion including stochastic variability , the anisotropic effects of the actomyosin cortical ring , and a boundary condition that does not impose an overall growth rate on the tissue . Second , and most important , we aim at elucidating the sufficient and required ingredients that endow the DV organizer with its features of functionality and robustness during the course of development as cell proliferation advances . To this end we test our model against the available experimental data and predict/quantify the effects when any of those components is missing . Our main conclusion is that the interplay between mechanical effects and the cell growth leads to the functionality and robustness of the growing DV organizer . Importantly , our results provide novel insights into the developmental mechanisms that drive the dynamics of the DV organizer and set a definition of the Notch fence model in quantitative terms and with regards to its sufficient and required contributions . Thus , we present evidence , both analytical and computational , that a distinctive regulation of the duration of the cell cycle is needed at the DV organizer for maintaining its features and stability , and that the cellular mechanical properties and the cleavage direction are coupled by the Hertwig rule . In addition , our in silico mutant analysis allow us to explore the role played by the differential affinity of cells at the compartments and the organizer and the actomyosin cable that develops at the DV boundary . The paper is organized as follows . The Results section first introduces the wild-type situation and shows that our modeling reproduces the dynamics and structure of a stable DV organizer that agrees with the available experimental data in terms of topological/size distributions [50] , cell division patterns [28] , [41] , cell response to ablation experiments [50] , and geometry adopted by ectopic organizers [28] , [29] . In addition , we also show that the DV organizer is robust with respect to mechanical perturbations like fast growing clones and parameter variation . Thus , our results unveil the mechanical and dynamical ingredients that are sufficient for explaining in a quantitative and predictive manner the growth of the DV organizer; in order to demonstrate that those are also necessary , we perform in silico experiments with lack-of-function mutants . In the Discussion section we elaborate the main conclusions that derive from our study and comment on their implications . Finally , in the Methods section , we flesh out our approach by describing the dynamical vertex model , and the implementation of the cell cycle and the cell division events . Therein we also detail the values of the parameters used in our simulations , the initial and boundary conditions , and the rules that control the cellular character .
Fig . 2 shows several snapshots that illustrate , from left to right and from top to bottom , the temporal evolution of a growing tissue under wild-type conditions ( see also Video S1 in the Supporting Information ) . Herein the term wild-type indicates that , as shown below , with the parameters used in our in silico experiments ( see Methods ) we are able to reproduce both qualitatively and quantitatively the dynamics of the DV organizer of an in vivo wild-type experiment ( see also robustness results below ) . The primordium comprises , as prescribed by the initial condition , two cell populations with differentiated mechanical properties: cells at the bulk of the compartments ( white ) and the DV organizer cell population ( red ) . As specified in the Methods section , the character of organizer cells can be lost depending on whether or not the cellular environment is able to maintain their Notch activity . Both compartments are identical in size and properties , i . e . dorsal/ventral regions satisfy the same dynamics . In that figure , the DV boundary has been highlighted by simply connecting the cell membrane edges of the DV organizer at opposite compartments . The initial phase of the cell cycle at each cell is taken as random and uncorrelated with respect to that of its neighbors . Yet , during the evolution of the growing tissue , correlations between the cell cycles of neighboring cells ( clustering ) naturally develop due to the division process ( see Video S2 ) . Our results about cell clustering are in qualitative agreement with experimental results that have shown that dividing cells are found throughout the entire disc as single cells or clusters of 2–10 neighboring cells [55] . Still , Milán and coworkers demonstrated that clustering is also driven by cell signaling that help mitotic cells to recruit neighboring competent cells , a problem that we have disregarded . Moreover , recent experimental results have revealed that the amount of clustering strongly depends on the experimental protocol [52] . Everything considered , a thorough quantitative comparison with experiments is difficult . Starting from the initial condition , the tissue is evolved dimensionless time units ( hours ) up to reaching a population of cells ( i . e . each cell undergoes , on average , three cell cycles ) . As detailed in the Methods section , the simulation follows the dynamics as determined by Eq . ( 3 ) , including simultaneous growth and division of cells . Visual inspection of Fig . 2 indicates that under these conditions the DV organizer grows straight ( see quantification below ) keeping a two-cell wide population and separating cells of opposite compartments . In addition , Fig . 2 reveals the robustness of the DV organizer scaling process as cell proliferation advances since it is able to cope with the stochastic variability of the cleavage orientation ( see Methods ) . Further robustness analyses to test the stability of the DV organizer can be done by means of mechanical perturbations and parameters variation ( see below ) . Besides these observations , different quantitative characteristics of the statistics of the growing tissue can also be extracted from the simulations in order to compare with experimental data . On the one hand , we analyze the cellular packing in relation to: i ) the histogram of the number of cell sides ( neighbors ) and ii ) the normalized average area distribution as a function of the number of cell sides . To this respect , we do not observe major differences when comparing cells at the organizer and at the bulk ( data not shown ) . On the other hand , we compute the distribution corresponding to the cumulative statistics of cell division orientations . As expected , in this case we do need to distinguish the populations from the organizer and from the compartments bulk . Results of these analyses are summarized in Fig . 3 . With regard to the packing statistics data , Figs . 3A–3B show the comparison of our in silico experiments with in vivo data of growing third instar wing imaginal discs from other researchers [50] . Trends and figures are well-reproduced; in particular we recover the preference for hexagonal coordination . The agreement indicates a reasonable accurate choice of parameters . As a possible source of the discrepancy , we note that the experimental figures derive from the analysis of static images while ours are obtained from the cumulative statistics of the tissue dynamics at different times . Interestingly , the histograms for cell division orientation reveal how the mechanical properties contrain the cellular cleavage . The angle of division is measured using as a reference an axis perpendicular to the DV boundary such that a null angle corresponds to cells that cleave orthogonally to the DV boundary ( see Fig . 3C ) . Fig . 3D reveals the differences between cells at the bulk ( right ) and cells at the DV organizer ( left ) . We recall that besides the Hertwig rule ( see Methods ) , no direction for the division is imposed . Consequently the histogram unveils a geometrical property of cells at the DV organizer: driven by the mechanical forces , those cells elongate along the DV boundary . Interestingly , this elongation is counterintuitively performed under the expense of positive line and cortical tensions ( see parameters values below ) . This persistence of oriented division is a distinctive attribute of organizer cells that is lost when the bulk compartment population is analyzed similarly . In that case the distribution is almost uniform thus indicating the lack of any spatial anisotropy in the division direction . With regard to the statistics of cell division orientation , it is worth mentioning an apparent discrepancy with the data from Major and Irvine [28] . There , a much more scattered division-orientation distribution for cells at the DV organizer , more uniform , yet different , to the cells at the compartment bulk , is reported . This disagreement can be attributed to the fact that their statistics actually refers to the orientation of the mitotic spindles of each cell relative to the nearest DV interface which may fluctuate significantly with respect to the actual cleavage direction . In fact , with respect to cells that belong to the organizer , our data are in better agreement with studies performed by Baena-López and coworkers that also quantify the post-mitotic allocation [41] . Still , there is some discrepancy that can be ascribed to the fact that these authors analyze longer developmental times . As for cells in the compartments , our data agree better with those of Major and Irvine since their analysis , as ours , focuses on regions close to the organizer as opposed to the studies of Baena-López et al . that explore longer developmental times and larger regions of the wing disc that contribute to the wing blade elongation . Further analysis of the model that allows for quantitative comparison with in vivo experiments addresses the cell response upon membrane ablation . Herein , we introduce these in silico experiments as a novelty . By checking the relaxation of the network of cell links , the implementation of these in silico experiments allow to establish the value of some parameters of the model . Unfortunately , there is a lack of systematic characterization of the damage that the laser wound induces on the cells cytoskeleton . The outcome of the experiments presents in fact high variability [39] , [50] . In our case , the ablation is implemented by suppressing a cell edge and , as a consequence , we eliminate for the corresponding cells and vertices all the energetic terms that depend on the suppressed edge , except for the elastic contribution due to the area . In particular , we remove the adhesion ( line tension ) contribution and further assume that the cortical ring is well attached to the cell membrane and consequently the wound modifies its properties by just removing the contribution of the wounded edge to its perimeter , while keeping intact the functionality for the rest of the ring ( Fig . 3E ) . The energy relaxation is then analyzed as in the in vivo experiments by monitoring the time-dependent separation of the two vertices whose common link has been removed , , with respect to the equilibrium distance . As shown in Fig . 3F , this relaxation is very close to exponential ( see also Video S3 ) , and thus characterized by a single time scale . Indeed the evolution is well fitted by the expression: ( 1 ) where , and is the relaxation time scale . We perform in silico ablations of different cell edges: those that belong to the DV boundary , edges from the organizer that do not belong to the DV boundary , and edges from cells of the compartments bulk . In all cases a complete relaxation is achieved at times of the order of seconds . In particular we obtain seconds . Moreover , edges from cells of the compartments bulk and edges from the organizer cells that do not belong to the DV boundary , behave dynamically in a similar manner ( data not shown ) . The main differences arise when comparing the displacement between edges that belong or not to the DV boundary . In that case , and in agreement with in vivo experiments in the context of the AP boundary [39] , [50] , the former develops an increased displacement due to a larger tension . Those differences can be characterized by the initial velocity of the expansion: and respectively , that is , a fold increase . We point out , that the comparison of our ablation data with those experiments in the context of the AP boundary is not feasible from a quantitative point of view since a different cellular environment is analyzed . Still , we expect our analysis to be predictive in that regard since an increased tension at the boundary has been experimentally reported in both cases . Additional analyses can be made in order to test that the components considered in our model are sufficient for reproducing and understanding the dynamics and growth of the DV organizer . In particular we focus on its stability against mechanical perturbations and on its dynamics when it is ectopically induced . The first test refers to the mosaic technique where clones of rapidly replicating cells are placed near the organizer ( Fig . 4A ) . In our in silico experiments those cells have the same mechanical properties that cells at the bulk , yet their duplicating time is of that of the latter . As a consequence of its growth advantage , the clone rapidly extends and exerts pressure over the neighboring organizer . This force is revealed by the bending of the DV organizer around the location of the clone ( see also Video S4 ) . Still , in agreement with experimental results [56] , the organizer grows intact and remains robust keeping the clone lineage restricted to its compartment . This fence-like picture contrasts with the situation where D and V compartment cells are placed in contact in the absence of an organizer ( see Fig . 5A ) . In that case , the compartments populations progressively mix and start to interdigitate as cellular proliferation advances . A second test has to do with another mosaic experiment . Clones at the compartment bulk driven by the ectopic expression of Delta in dorsal cells or that of Fringe in ventral cells are able to induce a functional organizer with fence-like properties as the wild-type one . Interestingly , the geometry of these clones evolves towards a roughly circular shape [28] , [29] . We mimic those experiments in silico by ectopically placing an organizer in the compartment bulk . Its initial shape is chosen to be not circular but “eight”-shaped while the rest of its properties are the same that in the wild-type situation . As shown by our simulations ( Fig . 4B , see also Video S5 ) , the system evolves in agreement with the experimental observations rounding up the organizer region , very much like fluid interfacial systems would do in order to reduce the surface energy due to surface tension . The biological plausibility of our proposal can be also evaluated from the perspective of the robustness with respect to parameters variation ( see Methods ) . This analysis addresses also the degree of freedom for choosing the wild-type parameter set . In this regard , of the parameter sets that lie within a distance develop a functional DV organizer: it does not break within the temporal window of interest . Parameter sets that lie within a distance produce a functional organizer in of the cases . Additionally , we evaluate the effect of varying the external line tension . If the latter is decreased by a , then the organizer develops robustly . On the other hand , if it is increased by a , then some simulations reveal that it threatens to break at some points when subjected to perturbations . Still , the dynamics is in all cases similar to the wild-type: we do not observe significant changes in the growth rate and the organizer keeps the compartments segregated ( see Fig . S2 ) . Therefore , our analyses reveals a robust mechanism for maintaining the mechanical stability of the DV organizer . In the previous section we have shown that our model includes ingredients that are sufficient to capture a repertoire of experimental observations involving both compartment cells and DV organizer cells in terms of the structure , stability , and dynamics of the latter . We now turn to check to what extent each of the components of the model is indeed required to account for the observed phenomenology . Our approach is based on different in silico experiments that mimic “lack-of-function” mutants by suppressing individual ingredients ( one at a time ) . We point out that while some of those “mutants” are not easily feasible , and consequently difficult to test , e . g . randomizing the cleavage direction at the DV organizer population , they do provide crucial information for analyzing their importance . The mutants considered are depicted in Fig . 5A–E , where a late snapshot of the growth after dimensionless time units ( hours ) is shown for comparison with the wild-type case of Fig . 2 ( see also Videos S6–S10 ) . While specific mutants reveal the role played by individual components , the best way to test the fence-like functionality of the DV organizer for restricting cell migration is to remove completely the organizer cells , e . g . a disc in the absence of Apterous activity [57] . Results in this direction are shown in Fig . 5A where it is evident that cells at opposite compartments intermix generating a finger-like pattern . Our first mutant test , Fig . 5B , refers to the case when the affinities ( line tensions ) between all cells , both bulk and organizer cells , are prescribed identical . Namely , organizer cells having the same affinity between them that cells at the bulk have , i . e . ( see parameters values below ) . However , we retain the differential values for both the line tension and the contractility terms that account for the actomyosin cable at the DV boundary . Under these conditions the DV organizer is largely disrupted and the fence-like properties disappear . The opposite situation is represented in Fig . 5C where we maintain the heterogeneous affinities between organizer and bulk cells , but we eliminate the actomyosin cable by suppressing its distinctive tensile term . Although the DV organizer is not wiped out as severely as in the previous case , the integrity of the organizer is noticeably weakened ( it lacks robustness ) , the DV boundary becoming less straight ( see quantification below ) , and threatens to be ruptured at several points under prolonged proliferation ( notice that at many locations the width of the organizer has been reduced to a single cell ) . This result is in agreement with experimental results showing that when the actomyosin cable is removed , the effectiveness of the compartment segregation is reduced [29] . The orientation of cell division is examined in Fig . 5D . This panel reproduces the result of the simulation when all cells in the tissue divide at random in terms of the cleavage orientation , i . e . cells do not follow the Hertwig rule . As expected from the results shown in Fig . 3D , this ingredient turns out to have a large impact on the stability of the organizer . Indeed , the DV organizer cannot maintain its stability and becomes easily disrupted . The role played by the distinctive cell cycle durations between organizer and bulk cells is finally analyzed in Fig . 5E . There , the differences in duration are removed and all cells are considered to have the same ( mean ) lifetime regardless of their lineage . In this case we observe that the DV boundary becomes very wiggly , and the DV organizer becomes wider . Note also that some cells cannot maintain Notch activity due to the widening . The structure and straightness of the DV boundary can be characterized by measuring the angle with respect to the DV axis for the cell edges that define it , as shown in Fig . 5F . In our simulations , in order to discard artifacts in the quantification , we check that and are not correlated quantities ( data not shown ) . Our data reveal that the straightest boundary corresponds to the case where the same affinities apply to all cells ( B ) . However , we recall that in this last case the DV boundary is easily disrupted and does not maintain its integrity in many regions . All other mutants lead to boundaries that are more wiggly than the wild-type case . Moreover , the histograms of angles indicate that in the wild-type situation the boundary preferentially organizes in a degrees zigzag configuration . The latter is also true for the cases C and D , yet showing a larger dispersion that contributes to their “waviness” . We notice that this -zigzag organization of the boundary is related with the predominant hexagonal topology of the tissue and not with the variability introduced in the cleavage orientation ( as large as degrees , see Methods ) as exposed by case D where the direction of division is random . The case where the organizer is removed , A , obviously displays the more outspread distribution with an almost uniform profile . In contrast , case B shows an uni-valuated distribution around zero degrees . Finally , case E reveals a wiggly organization of the boundary with an outspread angle distribution . Still , the latter is not uniform as in case A and indicates a preferential organization around degrees . Overall , we have shown that the distinctive mechanical properties of cells , the differences in cell cycle duration , and a cleavage criterion ( Hertwig rule ) are required elements for understanding the dynamics , structure , and stability of a robust growing DV organizer .
Herein we have proposed a dynamical vertex model to study the different dynamical ingredients that are both necessary and sufficient to understand , at a quantitative level , the mechanical maintenance of the DV organizer in the wing imaginal disc within the developmental time window we consider . Thus , we have shown that our model is able to reproduce , quantitatively when such data are available , the reported phenomenology in terms of packing statistics , division orientation , robustness against mechanical perturbations , relaxation dynamics upon membrane ablation , and the geometrical rearrangements of an ectopic organizer . Additionally , the validity of our model has been also corroborated from the viewpoint of the robustness with respect to parameters variation and to noise in the cleavage orientation . Importantly , we have presented evidence , both analytical and computational , that a distinctive regulation of the duration of the cell cycle is needed at the DV organizer for maintaining its features and that the feedback between the cellular mechanical properties and the cleavage orientation are coupled by means of the Hertwig rule . Whether or not these differences in cell cycle duration and the cleavage criterion are also necessary in the AP case is a subject of further research . Moreover , we have shown by means of an in silico mutant analysis , the effects of different contributions to the dynamics and regulation of this developmental structure and we have proposed a way to geometrically quantify the organization of the boundary . Our approach does not take into account either the molecular effectors or the interactions and pathways that underlie the specific functionalities that we have reviewed herein . Instead , we make use of an alternative procedure of analysis in order to evaluate their consequences in an effective way . The computational approach is then particularly valuable to the extent that it allows the implementation of tests that may not be easily feasible in vivo . All in all , our study provides a definition of the Notch fence model in quantitative terms and provides the elements that are both sufficient and required to keep and scale a robust and stable organizer with a well defined width over the course of development . There are crucial differences in our modeling approach with respect to previous implementations of the vertex model in the context of wing imaginal disc development [39] , [45] , [50] , [52] . First , we have included extra energetic terms in order to account for the asymmetry of actin-myosin expression at different cell edges . Second , our model includes a realistic , stochastic , dynamics of the cell cycle duration , its relation to the cell area growth , and a implementation of the Hertwig rule with cell-to-cell variability . Third , our modeling allows for simultaneous cellular growth and a continuous time description that permits to conduct , among others , in silico ablation experiments that provide relevant biomechanical information . Finally , in contrast to other approaches that assume periodic boundary conditions our simulations implement free-boundary ones that do not need to impose a rate for tissue growth . There are certainly consequences derived from this simulation scheme , namely , the pinching of the organizer and the packing properties ( e . g . orientation ) of cells at the periphery . The external border is pinched due to the extra tension and pulling exerted by the actomyosin cable that runs along the DV boundary . In addition , since the line tension is larger at the periphery than in the compartments , the best energetic strategy for cells at the periphery is to minimize their surface of contact with that external border and , consequently , cells elongate perpendicularly to it . Still , those effects are arguably only local and not relevant for the dynamics and overall behavior of the tissue . The fact that cells grow simultaneously lead to changes with respect to the topological properties observed in simulations using a quasistatic approximation . In particular , if cells are growing concurrently the appearance of a transient synchronization of the cell cycles , i . e . temporal correlations , leads to an increase of the tissue local pressure that is able to modify the distribution of cell areas and the statistics of cell neighbors . Indeed , at first sight any snapshot of the cell packing reflects this dynamics by exhibiting two classes of cells , those that are latent , with a characteristic minimal size , and those that are growing , with larger sizes . This feature can often be visualized in real epithelia and its precise quantification in the particular context of the wing imaginal disc is now possible using the Fucci technique [58] . The simultaneous growth of many cells explains why the quantities measured in our simulations fit reasonably well the experimental observations , even though the mechanical parameters lie in different regions of the phase diagram as compared to the case of fitting experimental data in quasistatic models ( see [50] ) . Incidentally , the values of the parameters reveal a complicated compromise , in some cases counterintuitive , between the energetic contributions . As a matter of discussion , we comment on the rule we implemented for maintaining the genetic profile of organizer cells ( see Methods ) . We effectively put into practice the fact that , during the temporal window of our interest , keeping Notch activity at organizer cells requires intercellular signaling between cells at the organizer and neighboring cells at the compartments via ligand-receptor binding [26] , [27] . In this context we have also considered the role of filipodia for robust signaling as recently reported [59] . If the rule for maintenance is ignored and the genetic profile of cells is simply inherited following a division event , then the obtained results , both for the wild-type and the mutants , are totally equivalent . In fact , the only case where some small differences can be seen corresponds to the case where the cell cycle durations are the same ( data not shown ) . This poses the interesting question about the relative importance of signaling versus mechanical and dynamical effects for the maintenance of the DV organizer during different developmental stages . At every developmental stage signaling is indeed fundamental . As a matter of fact , at early stages , mechanical and dynamical inputs do not seem to play a crucial role and signaling between compartments is the driving force for both , the formation and the stability , of the organizer [27] . Complementarily , our results suggest that , once the organizer has been established , the mechanical and dynamical contributions become fundamental for understanding how the organizer robustly deals with tissue growth . Altogether , we have shown that a modeling tool based upon a mechanical approach to the dynamics of tissue growth contributes to the quantitative and predictive understanding of the morphogenetic mechanisms that govern the evolution of the wing imaginal discs of Drosophila . This tool can feedback to in vivo experiments of the DV organizer in order to test our predictions and also be potentially implemented in other growing tissues where cell packing dynamics and biomechanical interactions are key elements .
The vertex model assumes that each cell can be represented by a discrete set of points: the apical vertices that define its characteristic polygonal-like morphology . Each vertex evolves independently and off-lattice driven by the biomechanical properties of the surrounding cells . It is assumed that the dynamics is purely relaxational , in the sense that it derives from the minimization of a ( time-dependent ) energy functional . Extending previous formulations , [50] , the energetic contribution of each vertex takes the form: ( 2 ) where the sums indexed by and run respectively over the cells and vertices sharing the vertex . The first three terms on the right hand side have been previously proposed [50] . The first term accounts for the elastic energy of cells , being proportional to the Young modulus , due to the difference between the actual cell area and the one that would have due to the cytoskeleton structure in the absence of the stresses associated to adhesion and cortical tension , . Note that the time-dependence of contains the information on the cell cycle ( see below ) and is what drives continuously the system out of equilibrium . The second term in Eq . ( 2 ) stands for the line tension , being the length of the edge connecting neighboring vertices and , and includes contributions from cell-cell affinities ( the action of molecules regulating the adhesion between cells , e . g . Armadillo ) and also cortical tension , with the parameter weighting these interactions . Finally , the last two terms model further contributions from the mechanical tension associated to the contraction of the actomyosin cortical ring . In this regard we distinguish two contributions . First , a term proportional to the squared cell perimeter , , that takes into account a global contractility effect . On top of that , we consider a local contribution that accounts for possible inhomogeneities of the contractile tension due to the accumulation of actin-myosin in specific regions of the ring as experimentally reported [28] , [29] . This last term is proportional to the squared edge length and herein will mimic the mechanical role played by the actin cable at the DV boundary . By neglecting inertial effects with respect to dissipation , we are led to consider an overdamped dynamics with a characteristic kinetic coefficient . Then , the equation of motion for vertex at position driven by the force may be written as , ( 3 ) The characteristic relaxation time is assumed to be much smaller than the average cell cycle duration , implying that the vertex configuration adapts almost immediately to the ( time-dependent ) local minimum of the energy . Notice that our scheme differs from the quasistatic approach used in previous studies where concurrent growth of cells is neglected [39] , [50] . Here we do not disregard the duration of the growing phase of the cell cycle . This makes our vertex model fully dynamical and more realistic even if is very small . We also note that a relaxational approach does not imply that cells behave as merely passive elements . For example , the cellular growth certainly involves an active cytoskeleton remodeling . However , the active contributions are either time independent or assumed to adapt sufficiently fast with respect to the time scale of the relaxation dynamics so that they do not need to be treated explicitly . In order to prescribe a biologically realistic evolution of the network topology ( i . e . network of vertices connections ) it is also necessary to include processes that may alter the cell neighbors environment , allowing for tissue plasticity: T1 recombination processes and T2 apoptosis/extrusion processes ( see [51] , [60] ) . T1/T2 processes are implemented in the in silico experiments when , where is a characteristic edge size defined as the length of a regular hexagonal cell with area , the minimum prescribed area of a cell ( see below ) . Finally , we set the external boundary of the tissue to be free to move with the same dynamics but with a larger positive line tension to keep the tissue sufficiently compact , close to a circular shape . The cell cycle duration is not deterministic but stochastic and depends on cell-autonomous processes and on the mechanical interactions with the cell local environment . Both are considered in our approach . Moreover , during the course of the cell cycle two distinct growth phases can be distinguished . On the one hand , there is a latent phase up to the middle of the cell cycle during which the cell does not grow . On the other hand , during the rest of the cell cycle , the cell grows such that the apical cell area increases in an approximately linear manner [52] , ( Bellaiche Y ( 2008 ) Private communication ) . We account for these observations as follows . Previous works have considered an internal clock , yet decoupled from the actual growth dynamics of the cell and its environment [52] . Herein , we first define an internal “clock” for each cell characterized by a temporal variable , , that measures the time elapsed since the beginning of the cell cycle . In addition we define the variable , such that is a deterministic time scale that accounts for a mean cell cycle duration in the absence of mechanical stress due to the cell local environment and is a random variable that accounts for the variability of cell cycle duration and that is assumed to follow an exponential distribution: The parameter controls the dispersion of the cell cycle duration , so that , with the above definitions , the average and standard deviations of are given respectively by , This approach for describing the cell cycle duration has been similarly hypothesized by other researchers and experimentally tested [61] , [62] , and reproduces the cell-age distribution such that , on average , the number of cells at the beginning of the cell cycle doubles that at the end of it following an exponential decay . Accordingly , here we choose . We then set the duration of the latent phase to be . As for the growing phase , we consider that the speed at which the apical area changes is the same for all cells such that , Thus , the time it takes for a given cell to double its preferred area is and its mean reads . Yet , the actual duration of the cell cycle that leads to the division event is prescribed to take place when the actual cell area , , reaches a certain threshold . Here we choose this threshold to be ( Bellaiche Y ( 2008 ) Private communication ) . The latter enforces the duration of the cell cycle to depend on the intra- and inter-cellular mechanical interactions . At completion of the cell cycle the cell is assumed to divide instantaneously and the cellular clocks of both daughter cells are reset . In our simulations the phase of the cell cycle in the initial configuration is randomly chosen for every cell . With respect to the cleavage direction , we implement the Hertwig rule that fixes a correlation between the longest axis of a cell and its division direction ( transversely to the former ) [63] . While there are several exceptions to this rule , e . g . Zebra fish gastrulation [40] , it is highly conserved among cell types and organisms . In particular , this rule holds for the epithelial cells of the wing imaginal disc of Drosophila; yet , a dispersion that can be as large as degrees has been registered ( Bellaiche Y ( 2008 ) Private communication ) , [64] . Recent studies suggest that this correlation persists in the wing during all its developmental stages [45] , [64] . Moreover , the influence of cell geometry on the positioning of the division plane has been thoroughly explored recently in other cell types ( sea urchin eggs ) [65] . This research has confirmed anew that the cleavage direction is set perpendicular to the longest axis of symmetry . In our in silico experiments , as a novelty , in order to determine the longest axis at division time , we evaluate the inertia tensor of the cell with respect to its center of mass assuming that a proper representation of the former is a polygonal set of rods , i . e . the cell edges . Upon diagonalization of the inertia tensor we obtain the principal inertia axes and subsequently the longest axis of the cell ( orthogonal to the direction along the largest principal inertia axis ) . Once the cleavage direction has been specified by these means we randomly ( bounded Gaussian ) implement a perturbation that may deviate the division axis up to degrees as aforementioned . This finally defines the two new vertices and consequently the new edge . During development , Notch activity eventually controls cell proliferation of cells at the organizer by arresting the G1-S cell cycle progression [66] , [67] . In fact , by late third instar the DV organizer and neighboring cells clearly define the so-called Zone of Non-proliferating Cells ( ZNC ) [66] . Thus , one may expect differences in the cell cycle duration of the organizer cells with respect to those at the compartments bulk [15] , [55] . While a precise quantification of this effect with respect to the DV organizer within the temporal window of our interest ( see below ) is missing , a simple , yet elucidating , geometrical argument allows us to estimate and predict those differences as follows . The organizer grows in one dimension since its thickness ( width ) remains constant , as opposed to the whole disc that grows roughly isotropically in two dimensions . Let us then suppose a growing disc . The area of such disc is then where and stand respectively for the number of cells and the average cell area . If is the average cell doubling time , , then and consequently the disc radius grows as . Therefore , if we want to make compatible the growth of the organizer within the disc we conclude that . That is , the average cell cycle duration of the cells growing in one dimension must be twice the average cell cycle duration of the cells growing in two dimensions . This geometrical argument suggest that , before the ZNC becomes specified , Notch activity at the organizer contributes by approximately doubling the cell cycle duration with respect to that at the bulk of the compartments and helps to maintain the straightness and the width of the growing DV organizer . By choosing the following characteristic scales of length , , and time , , Eq . ( 2 ) can be written in dimensionless form by defining the constants , , and . The assumption of fast mechanical relaxation ( driven by the energy functional ) compared to cell growth in our dynamical model thus implies . Whereas in the AP case a size difference between cells at the boundary and cells at the compartments has been reported [39] , so far there is no quantitative evidence that the same happens for the DV boundary . Consequently , herein we assume that for all cells , i . e . for all , ( in dimensionless units , ) . As shown in the Results section , we make use of different in silico experiments that we compare with in vivo experimental data for setting a meaningful value of the parameters: the histograms of the number of cell sides [50] , the cell area distribution [50] , the cell cleavage direction statistics [28] , [41] , and the dynamics of ablation experiments [50] . An initial guess of some parameters was taken from [50] and [39] . Still , the particularities of our problem from the biological and technical point of view leads to different , yet close , parameter values . In addition , the average duration of the cell cycle is also available , hours ( we use this value for cells at the bulk of the compartments and ∼20 hours for organizer cells . ) , which allows us to compute the area growing speed [68] . In order to test the effect of the mechanical interactions between cells for determining the actual duration of the cell cycle , in our simulations we also compute the latter and found hours for cells at the bulk and hours for cells at the organizer ( see Fig . S1 ) . Moreover , the ablation experiments provide a temporal scale for the mechanical relaxation ( see Results section ) : seconds [50] . By comparing these scales , we can state that the mechanical relaxation is approximately a hundred-fold faster than the growing speed . These scales can also be compared with the typical scale of gene expression processes . While the latter greatly depends on the problem under consideration , it is in any case faster than the cell cycle duration . For example , the timescale for the transcription and translation of Wingless is of the order of hours [69] . Consequently , with respect to the growing time scale of the cell , the gene expression dynamics can be adiabatically eliminated and remains stationary . Our initial condition assumes a set of regular hexagonal cells with a two-cells-wide stripe ( the fence ) specified as the DV organizer . We stress that this population corresponds to the signaling center from which Wingless is released . Yet , we do not consider the diffusion process but just its consequences with respect to the maintenance of the cellular character . In that regard , following a division event the “genetic” characteristics of the cell , in terms of its mechanical parameters , are inherited by its daughter depending on the cellular environment ( see below ) . The values of the mechanical parameters are chosen as follows . The global contractility parameter is the same for all cells in the primordium: . However , the line tension ( “affinity” ) between cells depends on the cell type [31] , [37]: , , and . Where the subscripts and denote organizer cells and cells at the compartment bulk respectively . Thus , in a compartment , the affinity between cells at the bulk is maximal , and the less favorable situation in terms of the line tension corresponds to the mixing between organizer cells and those of the compartment bulk . This in turn favors cell segregation . In addition , to account for the actin cable effects we distinctly prescribe: and for edges defining the DV boundary , i . e . edges shared by organizer cells of different compartments ( for all other edges in the in silico disc ) . As for the boundary condition , most authors that have implemented the vertex model have dealt with it using periodic boundary conditions . That simulation scheme implies that an overall tissue growth rate must be imposed ad hoc . Herein , we use free boundary conditions . Yet , the line tension of the in silico disc border , i . e . edges facing the exterior , is set to . The latter ensures that the tissue adopts a compact , roughly circular , geometry as seen in the wing pouch . Moreover , the main effect of the external line tension is to introduce an overall external pressure to the tissue , an effect that will always be present although one cannot easily quantify due to lack of experimental information . Yet , it is possible to deal with it effectively by including such external line tension . The aforementioned values of the parameters for the wild-type provides the best results in terms of the comparison with the available experimental data and the robustness to mechanical perturbations ( see Results section ) . We have also tested the robustness of our model with respect to parameters variation . That analysis is performed as follows . Taking as a reference wild-type parameters , , we generate for each parameter , , a new value at random ( uniform distribution ) that allows a variation , , up to ( up or down ) . We note that in our case , we exclude from this analysis the value of the line tension at the tissue periphery and the null value of for edges of the organizer cells that do not define the DV boundary , i . e . . That is , we allow parameters to vary . Each parameter set is then characterized by the vector . Consequently , the null vector corresponds to the wild-type situation . We notice that we check that the random sampling comprehensively explores the parameter space and that the obtained sets are scattered enough such that given any two set of parameters and then . In order to quantify the amount of variability for each parameter set , we compute the Euclidean distance to the wild-type condition , that is , the norm of the parameter vector . Consequently , in our analysis the maximum variation with respect to the wildtype condition is ( ) . We then run simulations with the new parameters set and check whether or not the organizer breaks . In total we performed simulations for different parameter sets . In addition , in order to analyze the effect of the external line tension , we perform simulations where we just vary this parameter up to . The dynamics of the DV organizer displays several , well differentiated , stages during development . Covering all of them is out of the scope of this study . We now briefly review some of these processes in order to better specify the temporal window we aim to describe . The DV boundary appears in middle stages of larval development [13] . In the middle of the second instar Apterous expression patterns the dorsal region of the wing disc [70] . Apterous first drives the Notch activation onset by promoting that dorsal/ventral Notch receptors get signaled by ventral/dorsal Delta/Serrate ligands [36] . This causes a two-three cells Notch activation stripe due to the receptor-ligand dynamics between opposing compartments . Herein we disregard this initial dynamics for activating Notch . Such incipient Notch activation is further amplified and stabilized as Wingless and Cut become expressed at boundary cells . Remarkably , Cut activity , which is evident by mid-third instar [36] , makes cells at the DV organizer refractory to Wingless signals that represses Notch activity via Dishevelled [27] , [71] . The latter removes ligands at the DV organizer cells and promotes polarized signaling: in order to maintain Notch activity DV organizer cells are forced to recruit ligands from cells adjacent to , but outside , the organizer [27] . In other words , at this stage , due to Cut activity , Notch activity is stabilized and sustained by cells of the compartment bulk adjacent to the DV organizer and not by ligands of opposite compartments as in the previous stage . We notice that in the past it was postulated that the stripe of early arising cut-expressing cells might be the barrier that separates dorsal and ventral compartments [72] . Moreover , the actomyosin cable that is regulated by Notch activity is evident at the beginning of the third instar and persists past the middle of the third instar [28] , [29] . Our modeling assumes this situation , i . e . early-mid third instar , as the initial timepoint and follows the dynamics during the next hours . Later , as development progresses , around 48 hours after the beginning of third instar , the boundary cannot be identified by F-actin staining , and two new cables start to develop at flanking cells [28] . In this study we do not consider these phenomena either . As mentioned above , our model accounts for two cells types with distinct mechanical and dynamical characteristics: cells at the compartment bulk , C , and cells at the DV organizer , O . According to the previous discussion , an O cell maintains its character as long as is in contact with a C cell . Namely , signaling from ligands at the bulk is necessary and sufficient for sustaining Notch activity at the organizer . While Notch-ligand signaling is supposed to happen between adjacent cells , recent studies have unveiled long range interactions . Thus , actin-based filipodia confer robustness to the Notch-Delta signaling mechanism and extend this interaction to cells that are not nearest neighbors [59] . Everything considered , in our model the O character of a cell is maintained if one , or more , nearest neighbor or next-nearest neighbor has a C character . Otherwise the O cell becomes a C cell: Notch activity is lost . In addition , we do not consider the reverse step: a C cell cannot turn into a O cell . Notice that the latter requires the initial onset of Notch activity established at previous developmental stages for eliciting Cut expression since Wingless signaling from O cells inhibits Notch activation in C cells . We develop our own code according to the prescriptions mentioned above . The algorithm for integrating Eq . 3 is a standard time-explicit FTCS ( Forward Time Centered Space ) . The code makes use of the parallel CUDA technology as we run our simulations in the computer GPU ( NVidia GeForce GTX 295 ) .
|
During development , tissues are shaped in order to form organs with specific functionalities . This process relies on mechanisms that control cell segregation and migration . These concepts are beautifully illustrated by the morphogenetic events that the imaginal discs of Drosophila undergo during metamorphosis . In particular , the cellular populations that will give rise to the dorsal ( D ) and ventral ( V ) parts on the wing blade are segregated and do not intermingle . The so-called organizer , a cellular population that becomes specified by the boundary of the D and V cellular domains , is responsible for this . Yet , how does the DV organizer robustly deal with the cellular growth in order to prevent cell mixing ? Moreover , how can the organizer be conveniently scaled as the tissue grows ? Herein we address these questions using a computational approach that takes into account the available experimental data . Thus , our study unveils the elements that are necessary and sufficient for understanding in a quantitative and predictive manner the dynamics , structure , and stability of a robust growing DV organizer: distinctive mechanical properties of cells , differences in cell cycle duration , and a well-defined cleavage criterion .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"physics",
"tissue",
"mechanics",
"developmental",
"biology",
"pattern",
"formation",
"biology",
"computational",
"biology",
"biophysics",
"biomechanics"
] |
2011
|
Dynamics and Mechanical Stability of the Developing Dorsoventral Organizer of the Wing Imaginal Disc
|
During chronic HIV infection , viral replication is concentrated in secondary lymphoid follicles . Cytotoxic CD8 T cells control HIV replication in extrafollicular regions , but not in the follicle . Here , we show CXCR5hiCD44hiCD8 T cells are a regulatory subset differing from conventional CD8 T cells , and constitute the majority of CD8 T cells in the follicle . This subset , CD8 follicular regulatory T cells ( CD8 TFR ) , expand in chronic SIV infection , exhibit enhanced expression of Tim-3 and IL-10 , and express less perforin compared to conventional CD8 T cells . CD8 TFR modestly limit HIV replication in follicular helper T cells ( TFH ) , impair TFH IL-21 production via Tim-3 , and inhibit IgG production by B cells during ex vivo HIV infection . CD8 TFR induce TFH apoptosis through HLA-E , but induce less apoptosis than conventional CD8 T cells . These data demonstrate that a unique regulatory CD8 population exists in follicles that impairs GC function in HIV infection .
In chronic HIV and SIV infection , viral replication is concentrated in B cell follicles in secondary lymphoid tissues [1–5] , although factors that promote this are not fully understood . Follicular helper T cells ( TFH ) , which reside in the secondary lymphoid follicles , are highly permissive to HIV [6] and exhibit anti-apoptotic properties [7 , 8] which likely contributes to viral persistence . We have previously shown that virus-specific CD8 T cells are present at lower frequencies inside the follicle compared to outside the follicle in HIV and SIV infection [2 , 9] , which may contribute to impaired viral clearance in the follicle . While CD8 T cells are present in the follicle , little is known about the function of these cells . We have previously reported that CD4 follicular regulatory T cells ( TFR ) are increased in number , exhibit heighted regulatory capabilities , and impair TFH proliferation and function in ex vivo HIV and in vivo SIV infection [7] . We hypothesized that follicular CD8 T cells may also have regulatory functions that further contribute to immune dysregulation in chronic HIV infection . Regulatory CD4 T cell populations can be readily identified based on expression of CD25 [10 , 11] , and Foxp3 [12] , their canonical transcription factor . Nevertheless , a consensus phenotype for CD8 Tregs has yet to be described . CD8 Tregs in the thymus and periphery of mice do not constitutively express Foxp3 [12] , and Foxp3-expressing CD8 T cells do not encompass CD8 Treg populations [13] . CD8 Tregs have been described in humans , but have limited defining characteristics , and most lack Foxp3 [14] . Thus , it is essential to demonstrate regulatory function with any CD8 Treg phenotype [15 , 16] . In mice , CD8 Treg function is dependent on B and T lymphocyte expression of Qa-1 , the murine equivalent of HLA-E , which binds to the TCR of CD8 T cells [17–19]; CD8 Treg function correlates with the affinity and duration of this interaction [18 , 20] . A specific subset of CXCR5hiCD44hi CD8 Tregs ( henceforth defined as CD8 TFR in this work ) was found to limit germinal center ( GC ) size and prevent autoimmune disease in mice [19] . The main targets of CD8 TFR are CD4 T cells [17] , specifically TFH [19] . In autoimmune-prone mice , CD8 TFR limit TFH expansion and autoantibody production [21] . CD8 TFR expressing CD122 ( IL-2Rβ ) in mice were also shown to inhibit CD8 T cell function through a mechanism involving IL-10 production , but not requiring TGFβ [22] . CD8 TFR differ from conventional CD8 T cells in their potent suppressive mechanisms and their dependence on IL-15 for function [19] . Importantly , cells with the CD8 TFR phenotype ( CXCR5hiCD44hi CD8+ ) have recently been identified in humans [23] . In the context of HIV infection there is limited evidence of CD8 Tregs . Stimulation of CD8 T cells isolated from HIV-infected patients with HIV peptides was shown to drive regulatory CD8 T cell function [24] . Suppressive function of HIV-specific CD8 T cells was further shown to be dependent on IL-10 production [25 , 26] . These HIV-specific CD8 T cells that produced IL-10 lacked both CD25 and Foxp3 , but were able to prevent HIV-specific cytolytic function [26] . Tim-3 has been previously shown to mediate CD4 Treg suppression [27] and impair virus-specific CD8 T cell responses [28] . CD8 T cells co-expressing PD-1 and Tim-3 have both exhausted effector function and produce high levels of IL-10 during chronic viral infection [29] . Upregulation of Tim-3 on CTL is strongly correlated with reduced cytotoxicity [30] and viral persistence [31] . Additionally , Tim-3+ CD8 T cells were increased in frequency and positively correlated with plasma viral load in progressive HIV infection [32] . Increased frequencies of CD8 T cells are also found in follicles of HIV-infected individuals compared to seronegative individuals [1] , but little is known about their functional properties . No studies to date have evaluated CD8 TFR in HIV or SIV infection and their potential role in HIV replication and persistence . In this study , we describe CD8 TFR phenotype and function in secondary lymphoid tissues during ex vivo HIV infection of human tonsils and in vivo SIV infection . We define CD8 TFR as CD3+CD8+CXCR5hiCD44hi cells and find that most follicular CD8 T cells in human and rhesus macaque secondary lymphoid tissues are CD8 TFR . CD8 TFR exhibit an enhanced regulatory phenotype in the context of ex vivo HIV infection and chronic SIV infection and are expanded in chronic SIV infection . In contrast to conventional CD8 T cells , CD8 TFR are able to impair TFH effector function via Tim-3 and inhibit GC B cell IgG production . Further , CD8 TFR modestly reduce HIV replication in TFH and induce less TFH apoptosis than conventional CD8 T cells in HIV infection ex vivo . Induction of TFH apoptosis by CD8 TFR is dependent on HLA-E . This work provides the first functional analysis of follicular CD8 T cell populations in the context of lentiviral infections and suggests that CD8 TFR contribute to impaired TFH function and GC dysfunction during HIV infection .
First , we determined the fraction of follicular CD8 T cells within human tonsils that exhibit a CD8 TFR phenotype two days following spinoculation with X4- or R5-tropic GFP reporter virus or mock spinoculation . CD8 TFR are defined as viable CD3+CD8+CXCR5hiCD44hi cells and conventional CD8 T cells ( CD8 conv ) defined as all other CD3+CD8+ cells as shown in representative flow cytometry plots in Fig 1A . The percentages of CD8 TFR are not significantly altered 2 days after HIV spinoculation compared to mock-spinoculated cells ( Fig 1B ) . Using flow cytometry counting beads to show the average of 3 experiments , we confirmed that CD8 TFR did not increase numerically in the context of ex vivo HIV infection ( S1 Fig ) . Additionally , we confirmed that percentages of CD8 T cells that are CD8 TFR do not differ between day 0 , prior to any spinoculation or cell culture , and day 2 following mock spinoculation and 2 days in culture media ( S2 Fig ) . Next , we analyzed CD8 regulatory and exhaustion molecules ( Tim-3 , CD122 , IL-15R , and IL-10 ) , the CD4 Treg mediator GITR [33] , and perforin expression on CD8 TFR and CD8 conv in the of context HIV infection ex vivo . First , there are no significant differences in CD122 , IL-10 or perforin expression in CD8 TFR between day 0 and day 2 in mock-spinoculated cultures ( S2 Fig ) . Tim-3 is expressed on a significantly greater percentage of CD8 TFR compared to CD8 conv in all culture conditions ( i . e . , mock- , X4- , and R5-spinoculated cultures ) and there are no differences related to HIV infection ( Fig 1C ) . Perforin expression is lower in CD8 TFR compared to CD8 conv in all culture conditions , and is not altered by spinoculation ( Fig 1D ) . The percentage of IL-10+ CD8 TFR is significantly higher following both X4- and R5-spinoculation compared to mock-spinoculated cultures , and is significantly higher than the percentage of IL-10+ CD8 conv in each condition ( Fig 1E ) . X4- and R5-spinoculation leads to significantly higher percentages of IL-15R+ CD8 TFR compared to mock spinoculation , and there are greater percentages of IL-15R+ CD8 TFR compared to IL-15R+ CD8 conv after HIV spinoculation ( Fig 1F ) . The percentage of GITR+ CD8 TFR does not increase after HIV spinoculation , but the frequency of GITR+ CD8 TFR is greater than the frequency of GITR+ CD8 conv in all conditions ( Fig 1G ) . CD122 is expressed on a greater percentage of CD8 TFR after X4-spinoculation compared to mock-spinoculation , and is expressed on a greater percentage of CD8 TFR compared to CD8 conv in all conditions ( Fig 1H ) . There are not high levels of or significant differences in expression of the exhaustion molecule PD-1 or the inhibitory CD4 Treg receptor CTLA-4 between CD8 TFR and conventional CD8 T cells ( S3A Fig ) . Further , a majority of CD8 TFR have an effector memory as opposed to central memory profile , and express more IL-10 , CD122 , and GITR and less perforin than conventional CD8 T cells regardless of effector or central memory cell designation ( S3B Fig ) . Downregulation of CCR7 , the extrafollicular retention molecule , combined with upregulation of CXCR5 , the follicular homing molecule , is crucial for migration of CD4 T cells into follicles [34] . It has previously been shown in human tonsils that CXCR5+CD8+ T cells were located in the follicles of human tonsils by immunofluorescent tissue staining [35] . We determined the phenotype of CD8 T cells expressing the follicular homing phenotype CXCR5+CCR7- after mock- , X4- , and R5-spinoculation of human tonsil cells . A median of more than 90% of CXCR5+CCR7- cells are CD8 TFR in all conditions whereas only a minority of CD8 conv exhibit a follicular homing phenotype ( Fig 2A ) . We further analyzed CCR7 expression on both CD8 TFR and CD8 conv . The percent of CCR7 positive cells is significantly higher in CD8 conv than CD8 TFR ( Fig 2B ) . We next investigated whether CD8 TFR are able to suppress the GC reaction by inhibiting TFH proliferation and cytokine production , B cell function , or both . T cell proliferation is dependent on IL-2 , and IL-21 production by TFH is critical for B cell affinity maturation in the GC [36] . We therefore determined whether CD8 TFR inhibit IL-2 and IL-21 production by TFH . As whole tonsil cells were spinoculated for the data generated in Figs 1 and 2 , all sorted populations were spinoculated to mirror previous conditions and allow exposure to HIV . We find that CD8 TFR inhibit both IL-2 ( S4A Fig ) and IL-21 ( Fig 3A ) production by TFH in sorted X4- and R5-spinoculated cultures . This inhibition is dose-dependent on the ratio of CD8 TFR to TFH and does not occur when TFH are cultured with CD8 conv ( S4B Fig ) . To investigate potential mechanisms regulating IL-21 production in TFH , we blocked Tim-3 in 1:1 CD8 TFR:TFH co-cultures . Blockade of Tim-3 abrogates the effect of CD8 TFR and preserves the percent of IL-21+ TFH at levels similar to TFH cultured alone ( representative examples: Fig 3B , all data and 3C ) , but this is not observed in TFH and CD8 TFR co-culture with isotype control antibodies ( Fig 3B and 3C ) or TFH and CD8 conv co-cultures ( S4C Fig ) . There is not a consistent trend in alteration of TFH proliferation rates , as determined by proliferation dye dilution , in the presence of CD8 TFR when cultured at a 1:1 ratio ( S5 Fig ) . Next , we determined if CD8 TFR influence total IgG production in tonsil cultures . We cultured isolated GC B cells ( CD19+CD38+ ) with CD8 conv , CD8 TFR , and/or TFH at equal ratios , mock- or X4-spinoculated the cells , and stimulated them with CpG-B for 6 days . There is not increased IgG production in TFH and B cell co-cultures compared to B cells alone , likely due to the use of CpG-B that activates B cells directly through TLR9 . In X4-spinoculated experiments , adding CD8 TFR to B cells alone or to B cells and TFH leads to decreased IgG production , however , adding CD8 conv to B cells has no effect on IgG production ( Fig 3D ) . As HIV replication is highly concentrated within TFH ex vivo [6] and in vivo [1 , 4 , 5] , we next investigated the effect of CD8 TFR on HIV replication in ex vivo infected tonsillar TFH using GFP-reporter viruses . Both the percentage of GFP+ TFH and the MFI of GFP is modestly reduced when sorted TFH are co-cultured 1:1 with CD8 TFR ( Fig 4A ) following either X4- or R5-spinoculation . Co-culture with CD8 conv , on the other hand , does not alter HIV replication in TFH ( S6 Fig ) . We next investigated potential mechanisms by which CD8 TFR reduce HIV replication in TFH . In mice , CD8 TFR inhibit TFH responses via Qa-1 [19] , the mouse equivalent of the HLA-E molecule . To determine whether this mechanism restricts viral replication in TFH , we added HLA-E blocking antibodies to co-cultures . Blocking HLA-E interactions in 1:1 CD8 TFR and TFH co-cultures has no effect on viral replication ( Fig 4B ) . HIV replication has also been shown to be inhibited by IL-10 in PHA-blasted PBMCs and cell lines [37] . Nevertheless , we do not observe differences in viral replication in co-cultures of CD8 TFR and TFH with addition of IL-10 neutralizing antibodies ( Fig 4C ) . However , inhibition of HIV replication in TFH by CD8 TFR is contact-dependent , as inhibition is not observed when cells are separated by transwell membranes in 4 separate experiments ( Fig 4D ) . As CD8 TFR are located within the follicle , their ability ( or lack thereof ) to kill TFH is important for understanding their role in the follicular concentration of HIV replication . In mice , CD8 TFR are able to recognize and kill autoreactive TFH via Qa-1 interaction [19] . Therefore , we wanted to determine if HLA-E , the human equivalent of Qa-1 , similarly allowed CD8 TFR to kill TFH in tonsil cultures . As tonsil cells are from HIV-uninfected subjects , we did not expect to see virus-specific killing and looked broadly at TFH apoptosis rates using Annexin-V . We investigated the rate of apoptosis in mock- , X4- , and R5-spinoculated TFH alone or in co-culture with CD8 TFR or CD8 conv . After cell populations were sorted , all cells were spinoculated , cultured together at 1:1 ratios for 2 days , and stained for Annexin-V to determine apoptosis rates of TFH . When cultured at a ratio of 1:1 with TFH , CD8 TFR induce less apoptosis in TFH than CD8 conv ( representative examples: Fig 5A , cumulative data and 5B ) . We find that the rate of apoptosis in TFH is reduced further by blocking HLA-E in 1:1 CD8 TFR and TFH co-cultures ( Fig 5A and 5B ) . Blocking HLA-E in CD8 conv/TFH co-cultures has no effect ( Fig 5A and 5B ) . The use of count beads during co-culture analyses shows that conventional CD8 T cells reduce the number of TFH in both X4- and R5-spinoculation , while co-culturing TFH with CD8 TFR reduces TFH numbers to a lesser extent ( Fig 5C ) . Next , we investigated the effects of SIV infection on the frequency of CD8 TFR in secondary lymphoid tissues in vivo . Using the same phenotype as in 1A for human tonsils , we define CD8 TFR as viable CD3+CD8+CXCR5hiCD44hi cells ( Fig 6A ) . We compared the percentage of CD8 TFR in disaggregated spleen and lymph nodes from chronically SIV-infected rhesus macaques to those from SIV-uninfected controls and find that percentages of CD8 TFR are increased a median of 4-fold during SIV infection compared to uninfected controls ( Fig 6B ) . Additionally , we determined whether the proportion of SIV-specific cells differed among CD8 conv and CD8 TFR . We find that the frequency of SIV-Gag specific CD8 TFR is similar to that of CD8 conv ( Fig 6C ) . There were positive , but statistically insignificant correlations between the percent CD8 TFR or percent tetramer+ CD8 TFR and viral loads ( S7 Fig ) . We determined the expression of key CD8 regulatory molecules on CD8 TFR in rhesus macaques as done with human tonsils in Fig 2 . In chronically SIV-infected animals , there is a significantly greater frequency of Tim-3+ CD8 TFR compared to CD8 conv ( Fig 6D ) . Perforin expression on CD8 conv is significantly greater in SIV-infected compared to SIV-uninfected animals , while CD8 TFR do not display this trend and have significantly less perforin than CD8 T conv in SIV-infected animals ( Fig 6E ) . The percentage of CD8 TFR that produce IL-10 is higher than CD8 conv in both SIV-uninfected and chronically SIV-infected macaques . Furthermore , the percentage of CD8 TFR that produce IL-10 is higher in SIV-infected compared to uninfected animals ( Fig 6F ) . The percentage of IL-10 producing CD8 conv is similar in uninfected and SIV-infected macaques ( Fig 6F ) . The percentage of Galectin-9 ( Gal-9 ) + CD8 TFR tends to be higher than the percentage of Gal-9+ CD8 T conv in SIV-infected animals , but differences were not statistically significant ( S8 Fig ) , while GITR is significantly increased on CD8 TFR in both SIV-uninfected and SIV-infected animals ( Fig 6G ) . As shown previously for human tonsils in Fig 2 , we determined the phenotype of CD8 T cells expressing the follicular homing profile CXCR5+CCR7- in lymph node and spleen of uninfected or chronically SIV-infected rhesus macaques . The vast majority of CXCR5+CCR7- cells are CD8 TFR ( median 89% of uninfected , 93% of SIV-infected ) whereas only a minority of CD8 conv exhibit a follicular homing phenotype ( Fig 7A ) . We further analyzed CCR7 expression on both CD8 TFR and CD8 conv . The percent of CCR7 positive cells is significantly higher in CD8 conv than CD8 TFR in both SIV-infected and uninfected rhesus macaques ( Fig 7B ) . Further , we performed immunofluorescent staining of lymph node and spleen tissue sections to determine the precise location of CXCR5+ CD8 T cells . As shown in representative images ( Fig 7C ) , the follicle was defined as CD20+ ( white ) and the number of CD8 T cells ( green ) expressing CXCR5 ( red ) were counted as either inside the follicle or in the extrafollicular zone . Most CXCR5+ CD8 T cells , a median of 91% in chronically infected rhesus macaques and a median of 93% in uninfected rhesus macaques , were found inside of the follicle ( Fig 7D ) . In light of findings above indicating that the vast majority of CXCR5+CD8+ T cells exhibit a CD8 TFR phenotype , it is reasonable to conclude that most cells with that phenotype reside in the follicle . Utilizing isolated CD8 TFR and TFH from chronically SIV-infected rhesus macaques , we determined if CD8 TFR regulated TFH function or apoptosis rates . Using disaggregated lymph node cells from 2 SIV-infected animals , we find that IL-21 production by TFH is inhibited when cultured with equal numbers of CD8 TFR ( Fig 8A ) but not CD8 conv ( S9A Fig ) . IL-21 production by TFH is not reduced in co-cultures treated with Tim-3 neutralizing antibodies in culture ( Fig 8A ) . The percentage of apoptotic TFH , determined by Annexin-V staining , is increased when TFH are cultured with equal numbers of CD8 TFR ( Fig 8B ) . When HLA-E interactions are blocked with neutralizing antibodies in TFH: CD8 TFR co-cultures , the percentage of Annexin-V+ TFH decreases ( Fig 8B ) but not when cultured with CD8 conv ( S9B Fig ) .
This is the first study to examine the role of CD8 TFR in the context of HIV and SIV infection . We determined that the majority of follicular CD8 T cells in human tonsils , as well as in secondary lymphoid tissues from rhesus macaques , exhibit a CD8 TFR phenotype . Of all the CD8 T cells expressing the follicular homing profile CXCR5+CCR7- , 90% or more of these cells are CD8 TFR in both human tonsils and rhesus macaque lymphoid tissues . Furthermore , their percentage is higher in lymph nodes and spleens of chronically SIV-infected rhesus macaques compared to uninfected macaques . A greater percentage of CD8 TFR express IL-10 and CD8 TFR express relatively lower levels of perforin compared to CD8 conv . In ex vivo HIV spinoculation and chronic SIV infection , CD8 TFR inhibit IL-21 production by TFH and IL-21 production is rescued by Tim-3 blockade . Further , CD8 TFR inhibit IgG production by B cells in ex vivo HIV infection . Additionally , CD8 TFR significantly , albeit modestly , reduce TFH permissivity to HIV ex vivo through contact-dependent mechanisms . Compared to CD8 conv , CD8 TFR induce a low level of TFH apoptosis . Blocking HLA-E interaction in CD8 TFR and TFH co-cultures further reduces TFH apoptosis rates . Collectively , these studies demonstrate that CD8 TFR constitute the majority of CD8 cells in B cell follicles and contribute to impaired GC function in lentiviral infections . Multiple factors may account for the inability of CD8 T cells to control HIV and SIV replication in the follicles . Heightened permissivity of TFH [6] combined with a paucity of virus-specific CTL in B cell follicles may foster virus replication at those sites [2 , 9 , 38] . Previous work in rhesus macaques has shown that CD8 is downregulated on SIV-specific cells entering the follicle [39] and CD8 downregulation leads to impairments of CTL cytokine production and proliferation [40] . It has been proposed that virus-specific CD8 T cells are exhausted in chronic infection [41 , 42] or that virus escape mutations prevent virus-specific CD8 T cells from recognizing infected cells [43 , 44] , but these hypotheses do not fully account for the follicular concentration of HIV replication . Here , we show that most of the follicular CD8 T cells in humans and rhesus macaques are CD8 TFR , CD8 TFR induce TFH apoptosis at lower rates than CD8 conv , and blocking HLA-E recognition further reduces CD8 TFR–mediated apoptosis in TFH . We also demonstrate that CD8 TFR produce a modest , but statistically significant reduction in HIV replication in human tonsil cells infected ex vivo . In the LCMV-infected mouse model , Leong et . al . [45] recently found that mice reconstituted with CXCR5+CD8 LCMV-specific T cells had 50% fewer LCMV-producing TFH compared to mice reconstituted with CXCR5-CD8 T cells . Additionally , this study showed that CXCR5+CD8 LCMV-specific T cells express less perforin and are less efficient killers of LCMV-infected cells than their CXCR5- counterparts ex vivo [45] . These findings are consistent with our findings in the ex vivo HIV infection model and in chronic SIV infection that CD8 TFR express less perforin than conventional CD8 T cells in both human tonsils and rhesus macaque lymphoid tissues , and that they induce modest reductions in HIV replication and TFH apoptosis ex vivo . In marked contrast , He et . al . [46] , reported profound reductions between 100- and 1 , 000-fold in splenic viral loads in the presence of CXCR5+CD44hiCD8 T cells compared to CXCR5-CD44hiCD8 T cells using the LCMV-infected mouse model . Reasons for the discrepancy between this study and that of Leong et . al . are not clear , as similar markers were used to define follicular populations of CD8 T cells in their studies , and these were the same markers that we used to define TFR CD8 populations in our study . An important area of future investigation will be to determine if CD8 TFR killing capacity can be boosted numerically and functionally to reduce follicular HIV replication in vivo . Mechanisms that underlie the modest reduction in HIV replication induced by CD8 TFR in our studies are not clear . This could be due to general immune suppressive mechanisms , such as lowering TFH activation levels , or alternatively through cell killing . Previous work has shown that CD4 Treg suppressed HIV replication in other CD4 T cell populations through contact-dependent inhibitory mechanisms rather than inducing cytotoxicity [47] . CD4 and CD8 regulatory cell subsets likely have complex roles in HIV infection , as evidenced by their ability to inhibit effector functions , and also reduce HIV replication . Further , CD8 TFR may promote the follicular HIV reservoir by inhibiting TFH IL-21 production , as it has been shown that ART combined with IL-21 therapy results in lower viral loads and lower intestinal cell-associated SIV DNA in rhesus macaques when compared to ART alone [48] . Determining at which point in the virus life cycle immune regulatory mechanisms inhibit viral replication could provide useful information to promote a balance of effector cell function and inhibition of HIV replication . As CD8 TFR display an ability to interfere with HIV replication ex vivo , future studies will be needed to determine if CD8 TFR inhibit viral replication or eliminate virus producing cells , as these both have distinct implications for HIV cure strategies . In this study , we show that CD8 TFR expand in chronic SIV infection . However , we do not see this expansion in our ex vivo tonsil model after 2 days of infection . A likely explanation for this discrepancy is that the expansion of CD8 TFR is a result of antigen-specific expansion in chronic HIV or SIV infection , and therefore not recapitulated in an acute infection model . This is supported by our data that show CD8 TFR are SIV-specific at a similar frequency as CD8 conv . We have previously reported that total levels of CD8 expression in lymph node follicles increase in chronically HIV-infected subjects compared to seronegative individuals [1] . The present study demonstrates that the increase of follicular CD8 T cells in chronic HIV infection is likely due to an expansion of CD8 TFR . Therapies aimed at generating protective virus-specific CTL in patients will need to verify that cytotoxic responses , as opposed to regulatory responses , are being produced in the follicle . An interesting question remaining is how CD8 TFR are generated during HIV and SIV infection . It has been suggested that persistence of antigen is the cause of functional impairment of HIV-specific effector responses [49] . TFH are expanded in HIV [4 , 50] and SIV [3] infection , upregulate HLA-E upon HIV infection [51] , and constant antigenic stimulation occurs in the follicle and GC [52 , 53] . It is possible that CD8 TFR are generated through contact with HLA-E on TFH as they enter follicular and GC regions . These interactions could convert effective CTL into CD8 TFR and allow for HIV persistence in the follicle . Further investigation as to whether CD8 TFR migrate into the follicle or are converted from precursors after migrating into the follicle will be an important aspect of potentially manipulating or preventing this response to promote cytotoxicity . It was shown that CD8 T cells from IL-15 knock-out mice lose suppressive capabilities [19] . Importantly , CD8 Treg from these mice were unable to suppress TFH number , GC expansion , and IgG production . Interestingly , expansion and cytotoxicity of CD8 T cells is impaired in IL-21 knockout mice , but the combination of IL-21 and IL-15 synergistically improves CD8 effector function [54] . Additionally , IL-15 has been shown to have therapeutic potential as an adjuvant in HIV vaccines [55 , 56] and in restoring CTL function [57 , 58] . Here , we find that CD8 TFR have elevated IL-15R expression and suppress TFH IL-21 production in HIV infection ex vivo . As CD8 TFR have access to HIV-infected TFH in the GC , it will be interesting to determine if the use of IL-15 and IL-21 in HIV therapy increases CD8 TFR cytotoxic function and promotes the elimination of HIV-infected TFH . In chronic HIV infection expression of exhaustion markers such as PD-1 [42] and Tim-3 [29] correlate with diminished cytotoxic capacity . The inhibitory cytokine IL-10 was specifically produced by Tim-3+PD-1+ CD8 T cells after stimulation with LCMV peptides , suggesting Tim-3 is a potential target to limit regulatory cytokine production and promote CTL function . We find that inhibition of Tim-3 prevents impairment of TFH function by CD8 TFR . Thus , Tim-3 blockade could provide a new mechanism to reverse CD8 regulatory functions and perhaps boost TFH function . Further studies addressing the inhibitory mechanisms of CD8 TFR and their role in TFH impairment would be useful to determine how to boost the high quality response of TFH to HIV vaccines . Previous work in mice showed that CD8 TFR do not directly suppress antigen-specific antibody production [19] , however CD8 T cells have been shown to directly suppress non-specific antibody responses in other models [59] . A previous study found that both CXCR5-CD8+ and CXCR5+CD8+ human tonsil cells supported IgG production by CD19+ B cells in uninfected , unstimulated conditions [35] . Here we find that CD8 TFR are able to directly suppress total IgG production by B cells in a stimulated , HIV-spinoculated culture . This is not observed when B cells are cultured with CD8 conv , indicating a unique effect of CD8 TFR on GC B cells . A previous study using human tonsil cells showed that CD4 Treg were also able to suppress IgG production by B cells [60] . Thus , the mechanism for direct suppression of B cells could be a general feature of all regulatory T cell subsets , through common expression of Tim-3 or GITR , or a mechanism unique to CD8 TFR such as sequestration of IL-15 since IL-15 knock-out mice have been shown to produce less IgG [61] . Further , it is unknown at this time if CD8 TFR can suppress antibody production elicited by TFH , but it is likely as CD8 TFR strongly inhibited IL-21 production by TFH . Future studies to determine whether CD8 TFR suppress HIV-specific antibody responses are warranted . CD8 TFR suppression of B cell function could potentially account for why HIV-infected individuals have poor responses to T cell-independent vaccines , such as the polysaccharide vaccines against pneumococcus [62] and S . typhi [63] . A better understanding of the role of CD8 TFR in generation of antigen-specific antibody responses could lead to innovative strategies to improve vaccine responses in HIV-infected individuals . Reduction of pro-inflammatory responses prior to high-dose SIV challenge prevented mucosal transmission [64] , suggesting regulatory immune cell function could be an important aspect of reducing HIV infection and replication . Interestingly , the induction of immune tolerance via CD8 Tregs was shown to have a protective role in rhesus macaques prior to SIV challenge [65] . Specifically , non-classical MHC-Ib/E restricted CD8 T cells from SIV-immunized animals inhibited CD4 T cell activation and SIV replication within autologous CD4 T cells infected ex vivo , but only if the CD8 T cells were added within 48 hours [65] . We similarly observe that CD8 TFR reduce HIV replication in TFH ex vivo after 2 days . Taken together with our results that CD8 TFR induce less apoptosis in TFH than conventional CD8 T cells , we hypothesize that CD8 TFR would limit cellular activation and HIV replication within TFH on a per cell basis whereas CTL would reduce the number of HIV-infected TFH . However , these data were obtained in an ex vivo infection model and lack HIV-specific responses , so further studies in vivo would be necessary to determine if CD8 TFR and CTL differentially affect the follicular HIV reservoir . Although we observe only modest reductions in HIV replication in tonsil cells infected ex vivo , our data supports the notion that inflammation and cellular activation promote HIV infection and replication . However , regulation of infected cells may have drawbacks , as opposed to benefits prior to infection . An in-depth analysis is necessary to determine if immune regulatory activity on infected cells is inhibiting replication pre- or post-integration and if this promotes formation of the latent HIV reservoir . In this study , we find that most follicular CD8 T cells are CD8 TFR and they potently impair TFH and GC B cell responses . The heightened regulatory function and relative lack of cytolytic potential of CD8 TFR could contribute to viral persistence in the follicle and impairments of humoral immunity that are characteristic of HIV and SIV infection . This is a novel mechanism of regulation of humoral immunity and remains to be explored in the context of other human diseases . Development of therapies that block CD8 TFR interaction with TFH and GC B cells could lead to novel approaches to improve the quality of antibody responses in HIV infection .
Human tonsils were obtained from the Colorado Children’s Hospital ( Aurora , Colorado , USA ) following routine tonsillectomy from individuals at low risk for HIV infection . Use of tonsil specimens for these studies was reviewed by the Colorado Multiple Institutional Review Board and determined to not constitute human subjects research ( COMIRB approval no . APP001-1 ) , in accordance with guidelines issued by the Office of Human Research Protections ( http://www . hhs . gov/ohrp/policy/checklists/decisioncharts . html ) , and consequently , informed consent was not required . All research involving human subjects conformed to the principles set forth in the Declaration of Helsinki and was approved by the Colorado Multiple Institutional Review Board . Rhesus macaques were cared for according to the guidelines of the Animal Welfare Act and the NIH for housing and care of laboratory animals . Animal experiments were approved by the Institutional Animal Care and Use Committee of the University of Wisconsin ( IACUC; protocol G00632 ) . Procedures were performed to ensure that discomfort was limited to that unavoidable in the conduct of the research plan . Animals were housed at the Wisconsin National Primate Research Center ( WNPRC ) , which is accredited by American Association of Accreditation of Laboratory Animal Care ( Animal Welfare Assurance No . A3368-01 ) . Sedatives were applied as necessary for blood and tissue collections and analgesics were used when determined appropriate by veterinary medical staff . Animals were fed standard monkey chow twice daily . Pain , distress , animal behavior , food and drink consumption was monitored and adjustments were made as necessary . SIV-infected rhesus macaques were singly housed , but had visual and auditory contact with at least one social partner , permitting the expression of non-contact social behavior . Animals had access to more than one category of enrichment at WNPRC . The IACUC proposal included a written scientific justification for any exclusion from some parts of the enrichment plan . Research-related exemptions are reviewed at least annually . Lymph nodes and spleen were obtained from 6 SIVmac239-infected and 6 uninfected Indian rhesus macaques ( Macaca mulatta ) . Animals were infected either intravenously or intrarectally with SIVmac239 , and had been infected from 12 to 241 weeks ( median , 19 . 5 weeks ) at the time that specimens were obtained . Plasma SIV RNA concentrations ranged from 3 . 78 to 6 . 45 log10 copies/ml ( median , 5 . 45 log10 copies/ml ) , and CD4+ T cell counts ranged from 291 to 422 cells/mm3 ( median , 362 cells/mm3 ) . Of SIV-infected animals , 2 were female and ranged in age from 7 to 8 years old . Of SIV uninfected animals , 2 were female and ranged in age from 10 to 23 years old . Tissues were either shipped overnight on ice in cold RPMI 1640 and disaggregated , or disaggregated and cryopreserved at the Wisconsin National Primate Research Center and later shipped on liquid nitrogen to the University of Colorado . The HIV-1 NL4-3-based CXCR4 ( X4 ) -tropic green fluorescent protein ( GFP ) reporter virus NLENG1-IRES [66] and the CCR5 ( R5 ) -tropic GFP reporter virus NLYUV3-GFP [67] were used for tonsil cell infections . Virus stocks were prepared by transfecting 293T cells ( ATCC ) with either X4 or R5 plasmid constructs ( Effectene , Qiagen 301425 ) in complete DMEM ( DMEM + 10% FBS , pen/strep , and non-essential amino acids ) , collecting supernatants , and spinning at 800 x g to remove debris . Viral stocks were stored at -80°C prior to use . After disaggregation , 5 x 106 tonsil cells were spinoculated with either GFP reporter virus or a mock spinoculation with an equal volume of complete DMEM for 2 hours at 1200 x g at room temperature . Cells were washed to remove unbound virus and media , and cultured for 2 days at 37°C with 5% CO2 in RPMI with 10% FBS , L-glutamine , and pen/strep ( R10 ) at a density of 1 . 5 x 106 cells/mL . Cells were collected and immediately processed for analysis by flow cytometry . Cells were blocked for 20 minutes with 2% BSA in PBS at 4°C and then stained for 30 minutes at 4°C in the dark . The following anti-human conjugated antibodies were used: CD3-APCCy7-UCHT1 ( Tonbo 25–0038 ) , CD8-eVolve605-RPA-T8 ( eBioscience 83–0088 ) , CD19-FITC-SJ25C1 ( Tonbo 35–0198 ) , CD38-violetFluor450-HIT2 ( Tonbo 75–0389 ) , CXCR5-PE-MU5UBEE ( eBioscience 18–9185 ) , PD-1-APC EH12 . 2H7 ( BioLegend 329908 ) , GITR-PECy7-eBioAITR ( eBioscience 12–5875 ) , CD44-APC-IM7 ( Tonbo 20–0441 ) , Tim3-APC-344823 ( R&D FAB2365A ) , IL-15R-FITC-eBioJM7A4 ( eBioscience 11–7159 ) , CD122-V421-Mikβ3 ( BD 562887 ) , PD-1-APC-EH12 . 2H7 ( BioLegend 329908 ) , CTLA-4-PECy5-BNI3 ( BD 561717 ) , CD62L-PE-SK11 ( BD 341012 ) , and CCR7-PECy7-3D12 ( BD 557648 ) . The same antibody panel was used for rhesus macaque cell staining with the exception of CD3 SP34-2 ( BD 557757 ) . All analyses were performed on Ghost Dye 510 ( Tonbo 13–0870 ) negative cells . Cells were fixed with 2% paraformaldehyde . Fresh human tonsil cells were typically 70–90% viable after culture and cryopreserved rhesus macaque cells ranged from 40–80% viability after freeze/thaw . All antibodies were used at one test per 106 cells . Data were acquired on a custom LSR II flow cytometer ( Serial # H47100196 , BD Immunocytometry System , San Jose , CA ) with BDFACS Diva ( v6 . 1 ) and with a configuration of 6 filters ( 755LP , 685LP , 670LP , 635LP , 600LP , 550LP , and 505LP ) on a blue laser ( 488 nm ) , 6 filters ( 750LP , 690LP , 635LP , 595LP , 505LP , and 450/50 ) on a violet laser ( 405 nm ) , and 3 filters ( 755LP , 685LP , and 670/30 ) on a red laser ( 633 nm ) . FCS files were analyzed using FlowJo ( v10 . 7 , Tree Star , Ashland , OR ) . Briefly , biotinylated MHC class I monomers were loaded with peptides ( NIH Tetramer Core Facility ) and converted to MHC tetramers with APC streptavidin ( Prozyme PJ27S ) . The MHC class I monomer Mamu-A*001:01 molecule loaded with SIV Gag CM9 ( CTPYDINQM ) peptides was used . Cryopreserved disaggregated lymphoid tissue cells were thawed , and 1–2 x 106 disaggregated cells were resuspended in 100 μL of tetramer staining buffer ( 5% fetal bovine serum in PBS with 0 . 06% sodium azide ) and incubated with APC-labelled Gag CM9 tetramer concurrently with CXCR5-PE-MU5UBEE , CD8-eFlour605-RPA-T8 , CD44-eFluor450-IM7 , CD3-APC Cy7-SP34-2 and Ghost Violet 450 viability in the dark for 40 minutes at room temperature . SIV-uninfected ( R05021 mesenteric lymph node , RhA578 inguinal lymph node , R05018 axial lymph node , and R05091 spleen ) and chronically SIV-infected ( R03111 spleen , Rh2123 spleen , R02017 inguinal lymph node , Rh2284 spleen , Rhax18 spleen , R02116 inguinal lymph node , and Ro1038 inguinal lymph node ) rhesus macaque lymphoid tissues were analyzed as follows . Four micron frozen tissue samples were fixed in 1% paraformaldehyde and stained with Rabbit anti CD20 ( Abcam , Cambridge , MA ) , Rat anti CD8 ( Bio Rad , Hercules , CA ) and mouse anti CXCR5 ( NIH NHP Reagent Resource ) followed by detection with AF647 Donkey anti rabbit , AF594 Goat anti Mouse ( highly cross absorbed ) and AF488 Goat anti rat ( highly cross absorbed ) ( Thermo Fisher Scientific , Waltham , MA ) . Sections were counterstained with DAPI and imaged on a Leica IMI6000 inverted fluorescent microscope . Ten 40X images were analyzed using Qwin ( Leica Microsystems ) by first defining the follicular region and counting CD8+CXCR5+ cells within each region and calculating the frequency of cells both in and out of the follicle . An adjacent section was stained for CD20 only and visualized using Vector HP Immpress and Vector Red ( Vector Laboratories , Burlingame , CA ) and the total follicular and extrafollicular area determined using Qwin . The percentage of CD8+CXCR5+ cells in the follicle was determined based on the frequency of cells within the follicular and extrafollicular regions and the total areas of each region . Human tonsil and rhesus macaque lymph node cells were sorted using a MoFlo Astrios EQ . Cells were sorted into TFH ( CD3+CD8-CXCR5+ ) , conventional CD8 ( CD3+CD8+CXCR5- ) , CD8 TFR ( CD3+CD8+CXCR5hiCD44hi ) , and GC B cell ( CD19+CD38+ ) populations . After sorting , all tonsil cell subsets were spinoculated with X4- or R5-HIV GFP reporter viruses and cultured in R10 . All cell populations were spinoculated to mirror whole tonsil cell cultures and allow for all cell populations to be exposed to virus . For measurements of TFH cytokine production , TFH were seeded at 1 x 105 cells in a 24 well plate and CD8 TFR were added at ratios of 1:1 , 1:10 , and 1:50 . After 2 days , TFH were stimulated to measure IL-2 and IL-21 by ICS . For blocking experiments , TFH and CD8 TFR or conventional CD8 were cultured at a 1:1 ratio in the presence of 500 ng/μL anti-Tim-3 ( Biolegend 345003 ) antibodies or 500 ng/μL anti-HLA-E ( Biolegend 342602 ) antibodies for 2 days . In transwell assays , 4 x 104 TFH were cultured in the bottom compartment and 4 x 104 CD8 TFR or conventional CD8 T cells were added to the upper compartment of 96-well permeable support plates ( Corning CLS3386 ) and cultured for 2 days . After 2 days of culture , tonsil cells were stimulated with 50 ng/mL of phorbol 12-myristate 13-acetate ( PMA , Sigma P8139 ) and 1 μg/mL of ionomycin ( Sigma I3909 ) in the presence of protein transport inhibitor containing monensin ( BD GolgiStop ) for 5 hours . Cells were then harvested , blocked and stained for surface markers as above , and then fixed and permeabilized using BD CytoFix/Cytoperm kit ( 554714 ) according to manufacturers’ instructions . Cells were then stained at 4°C for 30 minutes and analyzed for IL-2-PE-MQ1-17H12 ( BD 559334 ) , IL-10-eFluor450-JES3-9D7 ( eBioscience ) , IL-21-AF647-3A3-N2 . 1 ( BD 560493 ) , and perforin-FITC-Pf344 ( Mabtech 3465–7 ) . All cytokine analyses were normalized to a mock-spinoculated control that received monensin but was not stimulated . After sorting , 5 x 104 GC B cells and CD8 TFR TFH were cultured in the presence or absence of TFH at ratios of 1:1:1 . Cells were co-cultured in R10 for 6 days in 96 well plates and treated with either 2 . 5 μg/mL of CpG-B or were unstimulated . Media was collected and spun at 5 , 000 x g to remove cellular debris and stored at -80°C . ELISAs were performed using the total IgG kit ( Ready set go , eBioscience 88–50550 ) according to manufacturer’s instructions . Briefly , 96 well plates were coated overnight with anti-human IgG and culture supernatants were diluted at a 1:10 ratio for detection . Supernatants were incubated on pre-coated plates for 2 hours , washed , and incubated with HRP anti-IgG for 1 hour . After addition of HRP substrate , plates were analyzed on a plate reader at 450 nm and IgG calculated based on standards . TFH ( CD3+CD8-CXCR5+CD25- ) were sorted and stained with proliferation dye ( Cell Proliferation Dye eFluor670 , eBioscience 65–0840 ) at a concentration of 0 . 5 μM . In a 96-well plate , pre-coated with 5 μg/mL anti-CD3 ( Tonbo 40–0037 ) in PBS at 37°C for 2 hours , 104 TFH per well were cultured for 4 days in 200 μL R10 containing 2 μg/mL anti-CD28 ( Tonbo 40–0289 ) and 10 U/mL IL-2 with an equal number of sorted CD8 TFR ( CD3+CD8+CXCR5hiCD44hi ) or alone . At day 4 , cells were stained with viability dye ( Ghost Violet 450 , Tonbo 13–0863 ) and analyzed by flow cytometry . Comparisons of uninfected and SIV-infected rhesus macaque spleen or lymph nodes were performed using non-parametric Mann Whitney tests . Comparisons of tonsil cultures were performed using unpaired Mann Whitney or Friedman non-parametric tests . In direct comparisons of paired data , a paired Wilcoxon ranked sums test was performed to compare the two group medians of interest . Significance is denoted in each figure by asterisks , as * = p < 0 . 05 , ** = p< 0 . 01 , and *** = p < 0 . 001 . All statistical tests were performed with GraphPad Prism 6 .
|
HIV is a chronic infection and is never completely cleared from the body , despite successful antiretroviral therapy that reduces plasma viral loads to undetectable levels and restores CD4 T cell counts . While undetectable in plasma , HIV is able to hide in various niches throughout the body . One such niche are CD4 T cells residing in the follicles and germinal centers of secondary lymphoid tissues . The dynamics of these regions that lead to persistence of HIV-infected cells remain unclear . However , recent evidence strongly suggests that CD8 cytotoxic T lymphocytes , which are able to kill HIV-infected cells outside of these regions , are present at low numbers in follicles and germinal centers . Here , we further advance these recent findings by showing that the few CD8 T cells within the follicle have potent regulatory functions rather than conventional cytotoxic functions . Thus , the CD8 T cells entering these regions of HIV persistence not only fail to kill HIV-infected cells , but promote impairments in humoral immunity . These findings identify a new obstacle that must be taken into account to improve immune responses and clearance of HIV .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"hiv",
"infections",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"vertebrates",
"animals",
"mammals",
"retroviruses",
"throat",
"immunodeficiency",
"viruses",
"viruses",
"primates",
"animal",
"models",
"model",
"organisms",
"rna",
"viruses",
"amniotes",
"cytotoxic",
"t",
"cells",
"old",
"world",
"monkeys",
"research",
"and",
"analysis",
"methods",
"rhesus",
"monkeys",
"infectious",
"diseases",
"white",
"blood",
"cells",
"monkeys",
"animal",
"cells",
"medical",
"microbiology",
"hiv",
"t",
"cells",
"microbial",
"pathogens",
"siv",
"macaque",
"cell",
"biology",
"anatomy",
"tonsils",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases",
"lentivirus",
"neck",
"organisms"
] |
2016
|
Follicular Regulatory CD8 T Cells Impair the Germinal Center Response in SIV and Ex Vivo HIV Infection
|
Mycobacterial antigens are not exclusively presented to T-cells by classical HLA-class Ia and HLA-class II molecules , but also through alternative antigen presentation molecules such as CD1a/b/c , MR1 and HLA-E . We recently described mycobacterial peptides that are presented in HLA-E and recognized by CD8+ T-cells . Using T-cell cloning , phenotyping , microbiological , functional and RNA-expression analyses , we report here that these T-cells can exert cytolytic or suppressive functions , inhibit mycobacterial growth , yet express GATA3 , produce Th2 cytokines ( IL-4 , -5 , -10 , -13 ) and activate B-cells via IL-4 . In TB patients , Mtb specific cells were detectable by peptide-HLA-E tetramers , and IL-4 and IL-13 were produced following peptide stimulation . These results identify a novel human T-cell subset with an unorthodox , multifunctional Th2 like phenotype and cytolytic or regulatory capacities , which is involved in the human immune response to mycobacteria and demonstrable in active TB patients’ blood . The results challenge the current dogma that only Th1 cells are able to inhibit Mtb growth and clearly show that Th2 like cells can strongly inhibit outgrowth of Mtb from human macrophages . These insights significantly expand our understanding of the immune response in infectious disease .
Tuberculosis ( TB ) remains a major global threat because current interventions are unable to prevent or treat infection adequately . Mycobacterium tuberculosis ( Mtb ) is an intracellular pathogen that has evolved a myriad of effective evasion strategies to thwart host defence mechanisms . Due to increasing drug resistance , the continued impact of HIV co-infections and , more recently , the increasing impact of non-infectious co-morbidities in TB endemic areas , in particular obesity- associated type II diabetes mellitus , TB is unlikely to be conquered any time soon [1–5] . A major obstacle in designing more effective vaccination strategies against TB is our incomplete understanding of the human host response to Mtb , in particular the determinants that control protective immunity versus disease susceptibility [1–4] . This is e . g . illustrated by the unexpected failure of a recent vaccine trial using MVA85A , which was designed to boost BCG primed CD4+ Th1 cell responses , considered to be key to protection [6] . These results have led to a wide re-evaluation of current paradigms of the human immune response and protective host defence in TB , including the identification of major knowledge gaps . Current efforts to develop better TB vaccines include the development of subunit as well as live mycobacterial vaccines , and have generally aimed at inducing classical HLA class II and Ia restricted CD4 and CD8 Th1 cells . While canonical HLA class Ia and class II molecules are highly polymorphic , the HLA class Ib family contains only few allelic variants: 2 , 4 and 10 for HLA-E , -F and G , respectively [7] . Recently a novel coding variant for HLA-E was described , but this variation is unlikely to involve alternative peptide binding [8] . All described amino acid variations in HLA-E are located distant from the peptide binding groove , and in agreement with this , no differences in peptide binding capacities have been observed [9] . Physiologically , HLA-E is an interesting candidate antigen presentation molecule for new TB vaccine antigens . HLA-E is almost monomorphic , and its expression is enriched on Mtb phagosomes compared to classical class Ia family members , facilitating HLA-E peptide loading in Mtb infected cells [10] . Moreover , Mtb infected airway epithelial cells can also present Mtb antigens in HLA-E [11] . In addition , due to a mutation in the intracellular domain HLA class Ib family members are not sensitive to downregulation by HIV-nef proteins and thus should remain capable of presenting mycobacterial antigens during concomitant HIV-TB infection . Qa-1 , the murine equivalent of human HLA-E , is functionally important in mouse models of ( intracellular ) infectious diseases , underlining the functional contribution of non-classical class Ib restricted CD8+ T-cells to host defense . Pathogen specific Qa-1 restricted CD8+ T-cells can lyse infected target cells efficiently [12 , 13] . Antigens recognized from Salmonella typhimurium mimicked murine heat-shock proteins , resulting in potential recognition of stressed cells [12 , 14] . Antigen processing defects have been indicated as important triggers of Qa-1 restricted CTLs , supporting their role in immune-surveillance [13–15] . However , Qa-1-restricted CD8+ T-cells can also have regulatory activity , and can hamper efficient viral clearance in murine LCMV infection [16] . Qa-1 knockout have enhanced antiviral responses , resulting in reduced inflammation both in acute and chronic phases of the infection [16] . In humans , HLA-E restricted responses have been associated with effector responses towards infectious pathogens: HLA-E restricted responses were observed against cytomegalovirus ( CMV ) [17–19] , Salmonella typhi [20 , 21] , Mtb [22 , 23] and Epstein-Barr virus ( EBV ) [24] . Recognition of CMV and Salmonella typhi in the context of HLA-E resulted in production of IFNγ , as well as lysis of infected target cells by granule-dependent pathways [18 , 20] . HLA-E restricted responses to Salmonella typhi persisted long-term as they were detectable up to 2 years post-vaccination [21] . We have previously described the first peptides derived from Mtb that can be presented by HLA-E to human CD8+ T-cells . We demonstrated that some of these polyclonal CD8+ T-cells could suppress proliferation and cytokine production of Th1 cells , thus representing a subset of human CD8+ regulatory T-cells ( Tregs ) , whereas another subset of CD8+ T-cells had cytolytic activity towards BCG infected monocytes [23] . To unravel the function and specificity of human HLA-E restricted Mtb reactive CD8+ T-cells in more detail at the single cell level , we have performed an in depth analysis of CD8+ T-cell clones that recognize selected HLA-E presented Mtb peptides . While most Mtb reactive T-cells described have Th1 like functions , by contrast we find that HLA-E restricted Mtb reactive CD8+ T cell clones have unorthodox phenotypes , with many characteristics of Th2 cells , including the expression of the type 2 associated transcription factor GATA3 , type 2 cytokine expression , capacity to activate B-cells , and either suppressive or cytolytic functions . Importantly , these cells are present in patients with TB . Moreover , they are able to inhibit intracellular growth of Mtb , challenging the dogma that Th1 but not Th2 cells can contribute to mycobacterial growth control . These non-classical T-cells thus represent an important and novel ‘multifunctional’ subset engaged in the human immune response to infection .
Mtb derived peptides can be presented to CD8+ T-cells by the non-classical HLA class Ib molecule HLA-E [23] . To study this arm of the human T-cell response in more depth , we generated T-cell clones specific for either one of 2 Mtb derived peptides that are presented by HLA-E , using limiting dilution cultures . Thirty CD8+ T-cell clones from 2 independent donors were obtained , 16 of which could be characterized in more detail: 10 from donor 2 reactive against peptide #62 ( derived from Rv2997 , alanine rich dehydrogenase involved in secondary metabolites biosynthesis , transport and catabolism ) , and 6 from donor 6 reactive against peptide #68 ( derived from Rv1523 , methyltransferase involved in secondary metabolites biosynthesis , transport and catabolism ) . In all experiments the majority of these clones was included , guided by cell number availability . Each experiment included at a minimum 5 clones from each donor . All assays were performed on at least 12 independent T-cell clones in at least 3 independent experiments . The total set of combined data for all individual clones is given in S1 Table . Phenotyping by flow cytometry showed that 15 of the 16 clones had a CD3+CD8+ and 1/16 a CD3+CD4+CD8+ double positive phenotype ( Table 1 , S1 Table ) . The gating criteria are shown in S1 Fig . compliant with MIATA guidelines [25] , gate settings were determined using fresh PBMCs ( S1A Fig . ) and applied to our T-cell clones ( S1B Fig . ) . Twelve out of 13 T-cell clones tested expressed the αβ T-cell receptor ( TCR ) , whereas 1 clone expressed the γδ TCR ( Table 1 ) . Moreover , while CD56 expression was observed in 1/13 clones , all clones lacked CD94 , NKG2A , NKG2B and NKG2D , potential ligands for HLA-E ( Table 1 , S1B Fig . ) . Antigen-specific CD8+ T-cells induce the expression of CD137 ( 4-1BB ) upon TCR ligation by specific peptide/HLA complexes [26] . Following stimulation with their specific peptide in the presence of HLA-E , upregulation of cell surface CD137 was observed for 14/16 T-cell clones , but not following peptide presentation in the absence of HLA-E or following control peptide presentation , demonstrating specific T-cell activation by antigen presented via HLA-E ( Fig . 1A , 1B ) . A well-known early event following TCR activation is phosphorylation of the TCR associated signalling molecule ZAP70 . This occurs within seconds to minutes after TCR mediated recognition of peptide , does not require co-stimulatory signals and is one of the first cellular events following TCR activation . Analysing two independent ZAP70 sites ( ZAP70 Y292 and Y319 ) ZAP70 phosphorylation was found in 11/13 T-cell clones following peptide/HLA-E recognition , thus demonstrating TCR mediated T-cell activation in response to specific peptide presented by HLA-E ( Fig . 1C , 1D ) . Qa-1 restricted murine CD8+ T-cells have been associated with suppressor functions [16 , 27 , 28] , and we have previously reported a similar phenotype for human HLA-E restricted CD8+ T-cells [23] . Classical CD8+ T-cells are best known for their cytolytic capacity , and in line with this we previously showed that HLA-E restricted CD8+ T-cell lines not only had suppressive but also cytolytic functions [23] . To dissect these rather opposing functions in more detail we tested this at the single cell level in our panel of T-cell clones using highly standardized functional assays . Phenotypic analysis for both regulatory and cytolytic markers revealed mixed patterns . Most clones expressed markers associated with suppressor function , including CD25 , LAG3 and CD39 , however expression levels of these markers were highly variable ( Table 1 , S1 Table and S1B Fig . ) . In addition , several but not all CD8+ HLA-E restricted T-cells were able to exert dose-dependent suppression of the proliferative response of an independent reporter Th1 T-cell clone; Fig . 2A shows representative examples of clones with strong ( left ) , intermediate ( middle ) and no suppressor activity ( Fig . 2A ) . The suppression of proliferation correlated well with the inhibition of IFNγ secretion by the reporter Th1 clone . Although the level of suppression varied amongst the clones that were suppressive , 8/16 clones suppressed Th1 T-cell proliferation clearly and dose-dependently ( Fig . 2C left panel , S1 Table and S2 Fig . ) , in line with our previous results using bulk populations of T-cells . We next tested whether the HLA-E restricted CD8+ T-cell clones had cytolytic activity . Indeed , BCG infected monocytes could be lysed by several but not all clones ( Fig . 2B ) : 7 out of 12 clones tested had this cytolytic activity . Interestingly , 3 out of these 7 clones were able to lyse macrophages in the absence of BCG infection reproducibly ( S1 Table , specific lysis indicated between brackets ) . Moreover , 3 clones did not show any cytolytic nor suppressive activity , whereas 7 clones had either suppressive or cytolytic activity ( Fig . 2C and S2 Fig . ) . In addition to target cell lysis , and probably of more relevance in the control of intracellular Mtb , we assessed the capacity of our T-cell clones to inhibit outgrowth of intracellular Mtb . To this end we added the T-cell clones to Mtb infected macrophages and lysed them after 24 hours of co-culture after which Mtb was plated to assess the number of CFU . The majority ( 8/11 ) of the HLA-E restricted Mtb specific CD8+ T-cell clones had the capacity to inhibit Mtb outgrowth , whereas the 3 others did not affect ( nor promote ) Mtb outgrowth ( Fig . 2D ) . The level of Mtb growth inhibition differed amongst the clones ( Fig . 2C , right panel; S1 Table and S2 Fig . ) , and the magnitude of Mtb growth inhibition was found to correlate with the expression of perforin , perforin and granulysin or perforin and granzyme B ( but not granzyme B or granulysin in the absence of perforin , S1 Table ) expressed by the T-cell clones ( Fig . 2E ) . Taken together , the combined results suggest that the majority of the Mtb peptide specific HLA-E restricted human CD8+ T-cell clones has either cytolytic and Mtb inhibitory , or alternatively immune-regulatory functionality , whereas a minority of the cells was found to displayed dual functionality . The T-cell clones were further characterized by assessing RNA expression levels , cell associated ( surface ) markers and intracellular cytokines as well as secreted cytokines/ chemokines in supernatants . RNA expression analysis using dcRT-MLPA [29] confirmed CD3 and CD8 expression in the absence of CD4 ( Fig . 3A ) . Consistent with the expected phenotype of cloned T-cells they lacked CD45RA , but expressed CD45RO and CCR7 , compatible with an effector memory phenotype ( Fig . 3D , Table 1 ) . Moreover , as expected for CD8+ effector memory T-cells , cytolytic effector molecules granzyme A and B , granulysin and perforin were abundantly expressed ( Fig . 3B ) . The percentage of perforin positive cells was low compared to the percentage of cells expressing other cytolytic molecules , both at the level of RNA as well as by intracellular staining ( Fig . 3E ) , suggesting this to be a possible rate-limiting factor in cytolysis . Interestingly , when lineage determining transcription factors were assessed , we observed no Tbet ( TBX21 ) or RORC , but unexpectedly GATA3 expression was detected in all clones tested ( Fig . 3C ) . Transcription factor expression patterns and lineage determination were confirmed by flow-cytometry ( Fig . 3F , Table 1 ) . In addition , intracellular staining revealed eomes expression in 9/13 clones tested , in line with its critical role in the differentiation of effector CD8+ T-cells [30] ( Fig . 3F , Table 1 ) . Although FOXP3 mRNA was detected in 4/13 clones analysed , intracellular staining did not demonstrate any detectable protein expression ( Fig . 3C , Table 1 ) . Because GATA3 expression is associated with Th2 function , we next assessed the production of Th1 and Th2 family cytokines at the RNA and protein level , both using intracellular cytokine staining for flow cytometry and as secreted cytokines in supernatants . The HLA-E restricted CD8+ T-cell clones were capable of producing Th1 cytokines , including IFNγ ( ranging 70–8754 pg/ml following maximal stimulation using αCD3/28 beads; median 856 pg/ml ) and TNFα ( ranging 51–2826 pg/ml following maximal stimulation using αCD3/28 beads; median 393 pg/ml ) ( Fig . 4A ) , however , we only observed secretion of IFNγ when clones were stimulated maximally using αCD3/28 beads but not when stimulated with specific peptide or with human macrophages infected with BCG ( Fig . 4B , C ) . Thus , although these T-cell clones can be forced to produce Th1 cytokines under maximal , non-physiological stimulation and co-stimulation conditions , they fail to do so when stimulated under physiological conditions when seeing antigen ( including macrophages infected with mycobacteria ) ( Fig . 4C ) . All CD8+ T-cell clones evaluated were able to produce Th2 cytokines , as measured in supernatants following maximal stimulation with αCD3/28 . All clones produced IL-13 and IL-5 whereas 6/15 clones also produced IL-4 ( Fig . 4A , middle part ) . Secreted levels of IL-13 following maximum stimulation ranged from 1763–10 . 000 pg/ml , with a median production of 9153 pg/ml . IL-13 was already secreted at high steady state levels by the T-cell clones in the absence of additional stimulation , but in the presence of macrophages ( median 3471 pg/ml ) and these levels did not increase considerably by stimulation with either peptide loaded or BCG infected macrophages ( Fig . 4C , S1 Table ) . IL-4 and IL-10 levels varied among the clones: 6/15 clones produced IL-4 and 7/15 produced IL-10 ( mostly overlapping patterns ) whereas another 6 clones produced neither IL-4 nor IL-10 ( Fig . 4A ) . Interestingly , clones that secreted most IFN-γ and TNF ( following stimulation with αCD3/28 beads ) did not secrete IL-4 and IL-10 , suggesting some sub-dichotomy in Th1/ Th2 patterns for these CD8+ T-cell clones ( Fig . 4A ) . Intracellular cytokine staining confirmed IL-4 , IL-5 and IL-13 cytokine production by the CD8+ T-cell clones ( Fig . 4B , S1 Table ) . Cytokine production in response to specific peptide stimulation indicated an even stronger Th2 profile , with a virtual absence of any Th1 cytokine production ( Fig . 4C ) . Moreover , the same cytokine production patterns were also seen when cells were stimulated with BCG infected macrophages ( Fig . 4C , right panel and S2 Fig . ) indicating that natural processing and presentation of Mtb epitopes via HLA-E activates CD8+ T-cell type 2 cytokine production . Since the HLA-E restricted Mtb specific T-cell clones produced IL-5 and IL-13 , and some clones also produced additional IL-4 and IL-10 , we also investigated B-cell activation . Indeed , HLA-E restricted T-cell clones co-cultured with CD19+ B-cells induced increased expression of CD25 , CD80 , CD86 ( Fig . 5A ) and HLA-DR ( S1 Table ) on B-cells , indicating B-cell activation . Most HLA-E restricted Mtb specific CD8+ T-cell clones were in fact able to induce increased expression of CD80 , CD86 and CD25 on CD19+ B-cells ( Fig . 5B ) . Classical CD4+ and CD8+ T-cell clones did not activate B-cells ( Fig . 5C ) , suggesting that a specific property of HLA-E restricted CD8+ T-cells is the ability to activate for B-cells . Only the classical HLA class II restricted CD4+ T-cell clone R2F10 induced some B-cell activation , but this clone is known to produce IL-4 [31] . Since the HLA-E restricted CD8+ T-cell clones expressed various Th2 cytokines , we verified , using recombinant cytokines , that recombinant IL-4 induced the strongest upregulation of CD80 , CD86 and CD25 ( Fig . 5C ) . Activated B-cells produce and secrete IL-6 thus IL-6 in supernatants is considered an important hallmark of B-cell activation . Our T-cell clones did not produce IL-6 , even when stimulated maximally with αCD3/28 beads ( S1 Table ) . Supernatants collected from co-cultures of 11/13 HLA-E restricted CD8+ T-cells with primary B-cells contained IL-6 ( Fig . 5D ) , demonstrating B-cell activation . In contrast , supernatants from B-cells co-cultured with control T-cell clones did not contain IL-6 ( Fig . 5D ) . To further confirm the key cytokines involved in B-cell activation we added blocking antibodies to the co-cultures of HLA-E restricted Mtb specific T-cell clones that were capable of activating B-cells ( IL-6 > 200 pg/ml & IL-4 production by T-cell clone > 1000 pg/ml ) with B-cells and measured IL-6 levels in supernatants . Blocking antibodies to IL-4 inhibited IL-6 in supernatants of co-cultures , whereas antibodies to IL-5 or IL-13 did not inhibit B-cell secreted IL-6 , indicating that T-cell derived IL-4 is responsible for B-cell activation ( Fig . 5E ) . Finally , it was important to assess whether these HLA-E restricted Mtb specific T-cells were also present in the circulation of TB patients during active infection . T-cells recognizing these peptides were detectable in the blood of active TB patients directly ex vivo using HLA-E/peptide tetramers ( Fig . 6A ) . HLA-E tetramers containing peptide 62 were recognized by 13 out of 22 TB patients ( having more than 0 . 1% tetramer positive CD8+ T-cells ) with an average of 0 . 28% ( range 0 . 12–0 . 76% ) ( Fig . 6B ) . Moreover , HLA-E tetramers containing peptide 68 were recognized by 13 out of 18 TB patients , with an average of 0 . 32% of CD8+ T-cells ( range 0 . 18–1 . 3% ) ( Fig . 6B ) . Tetramer responses were absent in healthy uninfected ( PPD negative ) individuals , in accordance with our previous data [23] . We then studied the capacity of these peptides to stimulate cytokine production following overnight peptide stimulation of PBMCs of all TB patients ( Fig . 6C ) . In agreement with the above results from the T-cell clones , peptide stimulated PBMC from TB patients showed little or no TNF-α and IFN-γ production but produced significant levels of IL-4 and IL-13 , especially in the tetramer positive group ( Fig . 6D; using an arbitrary cut-off of tetramer responses > 0 . 10% as positive ) . Peptide 62 induced IL-4 and IL-13 responses by CD8+ T-cells that were significantly higher in the tetramer positive donors compared to the tetramer negative donors . Peptide 68 induced significantly increased IL-13 secretion by CD8+ T-cells of tetramer positive TB patients . For both peptides 62 and 68 all tetramer positive TB patients were capable of producing either IL-4 or IL-13 and for peptide 62 , 7 out of 10 patients produced both cytokines with levels higher than 0 . 1% cytokine producing T-cells . For peptide 68 the number of patients that was able to produce both cytokines upon the peptide stimulation was even as high as 6 out of 8 ( Fig . 6D ) . Taken together these findings demonstrate the presence of HLA-E restricted Mtb peptide specific T-cells in the circulation of TB patients . Moreover , they confirm the strong Th2 profile of the response against these peptides in active infection , in concordance with the phenotype of the HLA-E restricted , peptide specific CD8+ T-cell clones .
We here report that non classical , Mtb peptide reactive HLA-E restricted CD8+ human T-cells have an unorthodox phenotype: in contrast to classical Mtb induced CD4+ and CD8+ T-cells that are mostly Th1 type cells , these HLA-E restricted T-cells expressed GATA3 , predominantly produced the Th2 cytokines IL-4 , -5 , -10 and -13 , were able to exert either cytolytic or suppressive functions , and also provided B-cell help through IL-4 . Contrary to the dogma that Th2 cells may negatively impact on control of Mtb , these non-classical HLA-E restricted T-cell clones were able to inhibit intracellular Mtb growth , which correlated to their expression of the cytolytic molecules granzyme , perforin and granulysin . Moreover , patients with pulmonary TB had such ‘Th2-like’ non classical CD8+ T-cells in their circulation as visualised using specific Mtb peptide/HLA-E tetramers . Thus our results identify a new , non-classical , multifunctional T-cell subset which is engaged in the human immune response to infection , including during active TB . These results significantly expand our understanding of the human immune response to infection . Interestingly , all HLA-E restricted CD8+ T-cell clones we analysed expressed the Th2-associated transcription factor GATA3 and produced the Th2 cytokines IL-5 and IL-13 . Moreover , IL-4 and IL-10 were produced by half of the clones . GATA3 , classically known as transcription factor for Th2 differentiation , functions as a chromatin-remodelling factor for the IL-4 , IL-5 and IL-13 locus and can act as transcription factor for IL-5 and IL-13 genes [32] . Besides transcriptionally regulating expression of Th2 related cytokines , GATA3 also appears crucial for peripheral maintenance and proliferation of murine CD8+ T-cells [33 , 34] . Indeed , in murine LCMV infection and during alloantigen triggered graft versus host responses , GATA3 deficiency resulted in poor antigen specific T-cell proliferation [33] . In addition , GATA3 deficient mice have an impaired ability to kill tumour cells by CD8+ T-cells , indicating that GATA3 expression is required to mediate fully efficient cytolytic effector responses [34] . This agrees with our own observations here in that all of our cytolytic CD8+ T-cell clones expressed GATA3 . More importantly , the results document the involvement of GATA3 in anti-mycobacterial immunity for the first time , highlighting Th2 cytokine production , B-cell help and cytolysis as relevant GATA3 related functions in the immune response to mycobacteria . GATA3 expression levels within CD8+ T-cells seem a proper biomarker of immune dysfunction in patients with systemic sclerosis , a connective tissue disorder involving multiple organs [35] . As expected , the percentage of GATA3+ CD8+ T-cells correlated with the percentage of IL-13+ CD8+ T-cells , which are thought to play an important role in tissue fibrosis . Indeed in several skin disorders ( atopic dermatitis , psoriasis , sclerosis ) associated with fibrotic responses and inflammation , locally increased CD8+ IL-13 producing T-cells have been identified that seemed critical players [36–38] . Interestingly , the percentage of GATA3+ CD8+ T-cells was highest in the systemic sclerosis patients that also had interstitial lung disease , characterised by inflammation and fibrosis [35] . Moreover , also in human allergic asthma , IL-13 producing CD8+ T-cells isolated from the lung are increased and associated with airway obstruction [39] . Together these studies suggest that IL-13 producing CD8+ T-cells are important players in inflammatory disorders in the lung . Recently Th2 responses have been associated with TB in a number of studies . In zebrafish , mycobacterial infection induced Th2 gene expression signatures were associated with lower bacterial burdens , suggesting a contribution of Th2 immunity to control of mycobacterial infections [40] . In mice , Th2 responses during TB disease were initially considered to abrogate protection through inhibition of Th1 immunity , however , infection experiments in mice incapable of mounting Th2 responses have demonstrated that Th2 responses are not responsible for the inability to control Mtb infection [41] . However , Th2 immunity , such as IL-4 and IL-13 may be involved in disease related pathology , since overexpression of IL-13 in a murine Mtb infection model resulted in enhanced pathology , mimicking the human TB lesions closely [42] . In humans , mRNA levels for IL-13 were increased in Mtb infected lymph nodes , indicating local IL-13 expression in Mtb lesions , although the cellular source remained undefined [43] . BAL and plasma of patients with pulmonary TB contained increased levels of IL-4 compared to patients with other lung diseases , patients with moderate-advanced TB had higher levels of IL-4 compared to patients with mild TB disease [44] . Moreover , IL-4 producing T-cells have been found in the circulation of patients with pulmonary TB at diagnosis that disappeared rapidly following initiation of chemotherapy [45] . Here we now also show that patients with active TB have CD8+ T-cells that recognize Mtb derived peptides in the context of HLA-E , based on peptide/HLA-E tetramer staining . These results were further strengthened by peptide stimulation experiments that again confirmed the presence of a Th2 type response to these Mtb ligands . In contrast to previous hypotheses that postulated that Th1 but not Th2 were involved in controlling intracellular pathogens like Mtb , we show here that also Th2 cytokine producing CD8+ T-cells can actively lyse Mtb infected cells and , more importantly , limit intracellular Mtb growth . This is in line with a recent report from zebrafish in which the expression of Th2 cytokines was associated with lower bacterial burdens [40] . In patients , IL-4 producing T-cells were present at diagnosis and correlated with disease severity [44 , 45] , indicating that they are involved in the disease process , but unfortunately the anti-mycobacterial capacity of these cells was not reported . The anti-mycobacterial activity of our HLA-E restricted ‘Th2-like’ CD8+ T-cells suggests that these cells may contribute significantly to Mtb inhibition , and constitute a relevant part of the total immune effector repertoire against mycobacteria . In addition to the direct anti-mycobacterial activity of these ‘Th2-like’ CD8+ T-cells , we find that HLA-E restricted Mtb specific CD8+ T-cell clones can activate B-cells , with subsequent antibody production as additional component in the combat against Mtb . Recently , there has been a renewed interest in B-cells in TB mostly since B-cell related genes are specifically expressed during TB disease and change during successful treatment [29 , 46–48] . Experimental data have provided more direct evidence for the importance of B-cells in TB . Firstly , B-cell deficient mice appear more susceptible to TB [49] , secondly , B-cell follicle structures and activated B-cells have been found in granulomas of human and nonhuman primates infected with Mtb [50 , 51] . Thirdly , receptors for B-cell secreted immunoglobulins , and in particular the expression of the human Fcγ R1is a consistent and strong component of TB biomarker signatures [29 , 52–54] . In addition , recently a cytosolic Fc-receptor called TRIM21 [55 , 56] , was identified that can bind intracellular complexes of immunoglobulin bound to pathogen , resulting in subsequent immune activation and inflammation . These observations suggest that B-cells and/or Mtb specific immunoglobulins may play a hitherto unappreciated role in effector responses towards Mtb . The activation of B-cells by the Mtb specific HLA-E restricted CD8+ T-cells could be of potential benefit when utilizing this non-classical antigen presentation pathway in TB vaccination strategies , next to its limited allelic variation and its resistance to HIV-nef mediated downregulation . Mtb-specific HLA-E restricted CD8+ T-cells that produce IL-4 and IL-13 may home to the lung and participate in local immune responses during active TB . This could potentially lead to a variety of outcomes , including inhibition of Mtb by cytolysis and inhibition of intracellular Mtb growth , regulation of inflammation , tissue damage and fibrosis . However , Mtb specific HLA-E restricted CD8+ T-cells also include T-cells with potent immune-suppressing capacities , suppressing bystander ( CD4+ ) T-cell proliferation as well as effector cytokine production . These regulatory functions may dampen local inflammatory responses , particularly since these cells should have co-evolved within the former population of HLA-E restricted T-cells . The balanced induction of both functionally diverse populations may promote balanced inflammation [46] . Balanced immunity is thought to be critical to effectively control infection ( protective immunity ) as well as hyper- or hypo-inflammation ( pathogenic immunity ) , and needs to be tightly regulated by immune cells participating in host defence to the invading pathogen . Our finding that T-cells recognizing the same Mtb epitopes can have opposing , complementary functionalities in this respect may be relevant in achieving such a balance . It will be critical to further decipher how the balance between these complementary responses is determined before considering application of such peptides as vaccines . Thus , our data show that non classical human CD8+ T-cells recognize Mtb peptides in the context of HLA-E and have strong ‘Th2-like’ characteristics , can activate B-cells through IL-4 , and can either lyse infected target cells , inhibit intracellular Mtb growth or regulate inflammatory Th1 responses . These results reveal a novel , non-classical T-cell subset in humans which is engaged in the immune response to infection .
T-cell lines were generated from healthy anonymous bloodbank donors ( Sanquin Bloodbank , the Netherlands ) that had signed written informed consent for scientific use of blood products , donors 2 and 6 were the same donors as previously published [23] . Collection of PBMCs from TB patients collected at the University of Palermo , Italy was approved by the Ethical Committee of the University Hospital , Palermo , where the patients were recruited . The study was performed in accordance to the principles of the Helsinki declaration and those of the “Good Clinical Practices” , and all individuals gave written informed consent to participate . Peptide #62 ( RMPPLGHEL , Rv2997 , accession number O53244 ) and #68 ( VLRPGGHFL , Rv1523 , accession number Q50584 ) [23] were purchased from peptide2 . 0 Inc ( Chantilly , VA , USA ) . CD8+ T-cells from donor 2 recognized peptide #62 , but not #68 , whereas T-cells from donor 6 recognized peptide #68 but not #62 . In all experiments involving peptide recognition , specific and control peptide were compared . Control peptide in each case was the alternative peptide recognized by the other donor . Materials and methods are written in consensus with the most recent MIATA guideline ( minimal information about T-cell assays ) when applicable [25] . T-cell lines were generated by stimulation of PBMCs with peptide ( 10μg/ml ) for 2 weeks , followed by purification of CD8+ cells using magnetic beads ( Milteny Biotec BV , Leiden , The Netherlands ) . Lines ( 2x10e5 c/w ) were cultured in Iscove’s modified Dulbecco’s medium ( IMDM , Gibco Life technologies , Thermo Fisher Scientific Inc , Merelbeke , Belgium ) , supplemented with 10% pooled human serum and were restimulated in 96 well round bottom plates with irradiated ( 30 Gy ) pooled ( 5 donors ) PBMCs pre-pulsed with peptide ( 25 μg/ml , 5x10e5c/w ) , in the presence of IL-7 , IL-15 ( both 5 ng/ml , Peprotech , Rocky Hill , NJ ) and IL-2 ( 50U/ml , Proleukin , Chiron , Amsterdam , the Netherlands ) . Every other day cells were split and fresh IL-2 ( 100 U/ml ) was added . Cultures of purified CD8+ T-cells were incubated during 16 hours with fresh peptide pulsed feeder-cells , subsequently , cells were labelled with CD137-PE ( BD Biosciences , Erembodegem , Belgium ) , followed by incubation with PE beads ( Miltenyi Biotec BV , Leiden , the Netherlands ) and CD137+ cells were isolated . CD137+ cells were diluted to 0 . 3 cells per well and plated in 96 well round bottom plates containing 5x10e5 peptide-pulsed irradiated feeder cells in the presence of IL-2 ( 50U/ml ) . After two weeks of culture , growing clones were selected from the 0 . 3 c/well cultures and expanded as described above with alternating peptide pulsed irradiated feeders or T-cell expander beads . In total 30 clones were isolated , of which 16 randomly selected clones were analysed in detail . T-cell clones were stained for surface expression , intracellular markers or cytokines; live/dead stain ( Vivid fixable violet reactive dye , Invitrogen , Thermo Fisher Scientific Inc , Merelbeke , Belgium ) was used for all samples according to the manufacturer’s protocol . T-cell clones were further characterized in detail by cell surface staining directly from culture for CD3-PE-TexasRed ( Invitrogen ) , CD4-PE-Cy5 , CD8-HorizonV500 , CD94-PerCP-Cy5 . 5 , TCR-αβ-FITC or TCR-γδ-FITC , CD56 PE-Cy7 , CD16-BrilliantViolet605 , CD127-BrilliantViolet 650 ( all BD Biosciences ) , NKG2A-APC , NKG2C-PE , NKG2D-AlexaFluor700 ( R&D Systems , Abingdon , UK ) , or intracellular with fixation and permeabilization reagents ( ADG , ITK Diagnostics , Uithoorn , The Netherlands ) for CD3-Alexa700 , CD4-PE-Cy7 , CD8-HorizonV500 , CD27-PE , CD45RA-FITC ( all BD Biosciences ) , CCR7-APC-Cy7 , Tbet-BrilliantViolet605 ( Biolegend , ITK Diagnostics , Uithoorn , The Netherlands ) , RORC-APC , GATA3-PerCP-eFluor710 , FoxP3-PECy5 and Eomes-PE-C594 ( eBioscience , Vienna , Austria ) . Cytolytic molecules were assessed after a 24 hour stimulation with T-cell expander αCD3/28 beads ( Invitrogen ) by intracellular staining with rabbit-anti-human Granulysin ( kind gift of Dr . A . Krensky , Stanford , CA ) followed by Goat-anti-Rabbit-FITC , CD3-PE-Cy5 , CD4-PE-Cy7 , CD8-HorizonV500 , CTLA4-PE-C594 , CD25-APC-H7 , Perforin-PE , GranzymeB-AlexaFluor700 ( all BD Biosciences ) LAG3-Atto647 ( Enzo Life Sciences BVBA , Raamsdonksveer , the Netherlands ) , GranzymeA-PerCPCy5 . 5 ( Biolegend ) . Finally , cytokine profiles of the T-cell clones were analysed after addition of T-cell expander beads for 6 hours followed by 16 hours incubation with BrefeldinA ( 3 μg/ml , Sigma-Aldrich Chemie BV , Zwijndrecht , the Netherlands ) . Cells were stained for surface expression of CD3-PE-TexasRed ( Invitrogen ) CD4-PE-Cy5 , CD8-HorizonV500 , and intracellular for GATA3-PerCP-eFluor710 , TNF-PE-Cy7 , IL-2-BrilliantViolet605 , IFN-γ-AlexaFluor700 , IL-13-PE , IL-4-PE ( all BD Biosciences ) , IL-5-PE ( Biolegend ) , IL-10-APC ( Miltenyi ) and CCL4-FITC ( R&D Systems ) . T-cell clones ( 1x10e5 c/w ) were incubated with peptide loaded K562 cells ( 5x10e4 c/w ) with or without the HLA-E allele ( kind gift of Dr . E . Weiss , Ludwig-Maximilians-Universität , Munich , Germany ) [57] in multiple wells in a 96-well round-bottom plate . Stable surface expression of HLA-E is induced by 24 hour incubation at 26°C followed by peptide loading ( 20 μg/ml ) for 16 hours at 26°C and stabilization at 37°C for at least 2 hours prior to use . After 4–6 hours of co-culture , BrefeldinA ( 3 μg/ml ) was added and cells were incubated for an additional 16 hours before flow cytometric analysis was performed . Cells were harvested and stained for CD3-PE-TexasRed ( Invitrogen ) , CD4-PE-Cy7 , CD8-HorizonV500 ( BD Biosciences ) , followed by intracellular staining for CD137-PE-Cy5 ( BD Biosciences ) using fix/perm reagents ( ADG ) . Specific TCR triggering was assessed by phosphorylation of zap70 with Phosflow analysis . T-cell clones ( 0 . 5–1x10e6 cells ) were incubated in a 24-well plates for 5 minutes at 37°C in the presence of peptide pulsed MelJuSo cells ( 35000 cells/well , cell line was kindly provided by Prof . J . Neefjes , Dutch Cancer Institute , Amsterdam , the Netherlands ) . T-cell expander beads ( Invitrogen ) , were used to assess maximum zap70 phosphorylation . After incubation , T-cells were fixed ( BD lyse/fix for Phosflow buffer , BD ) for 10 minutes at 37°C , permeabilized for 30 minutes at 4°C with perm buffer III and stained for 1 hour at 4°C with the Phosflow reagents according to manufacturer’s protocol ( CD3-PE-TexasRed ( Invitrogen ) CD4-PE-Cy5 , CD8-HorizonV500 , Zap70-pY292 AlexaFluor647 , Zap70-pY319/Syk-pY352 –PE , all BD Biosciences ) . Cells were acquired on a LSRFortessa with Diva software ( v6 . 2 , BD Biosciences ) . Analysis was performed with Flowjo software ( v9 . 5 . 3 , Tree Star Inc , Ashland , OR ) . BCG ( Pasteur strain ) or Mtb ( H37Rv ) was grown in Middlebrook 7H9 medium supplemented with 10% ADC ( BD Biosciences ) , log phase bacteria were used for infection experiments . Multiplicity of infection was calculated based on determination of the number of viable bacilli per ml by plating serial dilutions of bacteria on Middlebrook 7H10 agar plates supplemented with 10% OADC ( BD Biosciences ) and counting of visible colonies after 3 weeks . Infections of monocytes , macrophages and adherent Meljuso cells were done at a MOI of 10 . HLA-E restricted peptide specific T-cell clones were tested for their ability to inhibit proliferation of a Th1 responder clone ( Rp15 1-1 ) as previously described [23 , 58–60] . Rp15 1-1 T-cells ( 1x10e4 c/w ) were cultured in a 96-well flat-bottom plate with irradiated ( 20 Gy ) , HLA-DR3 matched PBMCs as antigen presenting cells ( 5x10e4 c/w ) and 0 . 05–0 . 1 μg/ml of hsp65 peptide 3–13 , specific for the Th1 responder clone , in the absence or presence of HLA-E restricted T-cell clones ( 0 . 6–5x10e4 c/w ) . Proliferation was measured by [3H] TdR incorporation ( 0 . 5 μCi/well , Perkin Elmer , Groningen , the Netherlands ) after 96 hours . Cells were harvested with a 96-well Tomtec cell harvester ( Synchron , Etten-Leur , the Netherlands ) and counts per minute ( cpm ) were determined using a Wallac MicroBeta counter ( Perkin Elmer , Groningen , The Netherlands ) . Cytotoxic capacity of the T-cell clones was tested in a standard 51Cr release assay [23 , 58 , 61] . PBMCs from a HLA-A2 negative ( given the possible overlap in peptide binding profiles between HLA-E and HLA-A2 molecules [62] ) buffy coat were plated at 1 . 5x10e5 cell/well in a 96 well flat bottom plate for 5 days . Non adherent cells were washed away and cells were incubated with specific or control peptide ( 10μg/ml ) or were infected with live BCG ( MOI of 10 ) for 8 hours followed by incubation with 1 μCi 51Cr for 16 hours . The next day , cells were washed three times and T-cell clones were titrated on the target cells and incubated for 5 hours , followed by measurement of 51Cr release on a Wallac Wizard2 gamma counter ( PerkinElmer ) . Percentage specific lysis was calculated per well ( ( sample release/maximum release ) *100% ) . Macrophages were generated from CD14+ monocytes that were isolated from HLA-A2 negative buffycoat PBMCs with CD14 MACS beads ( Miltenyi ) and differentiated for 6 days in the presence of 50 ng/ml M-CSF ( R&D systems ) . Macrophages were harvested and seeded at 3x10e5 c/well in a 24 well plate for adherence . After 18 hours macrophages were infected with Mtb H37Rv from a log phase culture at a MOI of 10 for 1 hour followed by three washing steps with culture medium in the presence of gentamycin ( Lonza Benelux BV , Breda , the Netherlands ) ( 2 times with 30 μg/ml and once with 5 μg/ml ) . Infected cells were rested overnight , the next day T-cell clones were added at an E:T ratio of 5:1 in duplicate in the presence of specific peptide . After 24 hours of co-culture , the cells were lysed and serial dilutions were plated on 7H10 agar plates , supplemented with BBL Middlebrook OADC enrichment ( BD Biosciences ) for Mtb CFU determination . Colonies were counted after two to three weeks incubation at 37°C . Specific intracellular Mtb growth inhibition was calculated per experiment after substraction of the average variation in Mtb CFU in the absence of T-cells , all clones were tested in duplicate in 3–4 independent experiments using unrelated Mf donors and results of these experiments were averaged to obtain the percentage specific Mtb growth inhibition . HLA-E restricted T-cell clones ( 1x10e6 cells ) were lysed in TriZol reagent ( Invitrogen ) and RNA was isolated according to the instructions of the manufacturer . RNA was quantified using a Nanodrop ND-1000 spectrophotometer and diluted to 50 ng/μl for use in dcRT-MLPA . A dual-colour reverse transcriptase multiplex ligation-dependent probe amplification ( dcRT-MLPA ) assay was performed as described previously [29] . Briefly , for each target-specific sequence , a specific RT primer was designed , located immediately downstream of the left and right hand half-probe target sequence . Following reverse transcription , left and right hand half-probes were hybridized to the cDNA at 60°C overnight . Annealed half-probes were ligated and subsequently amplified by PCR ( 33 cycles of 30 s at 95°C , 30 s at 58°C and 60 s at 72°C , followed by 1 cycle of 20 min at 72°C ) . Primers and probes were from Sigma-Aldrich Chemie ( Zwijndrecht , The Netherlands ) [29 , 63] and RT-MLPA reagents from MRC-Holland ( Amsterdam , The Netherlands ) . PCR amplification products were 1:10 diluted in HiDi formamide containing 400HD ROX size standard and analyzed on an Applied Biosystems 3730 capillary sequencer in GeneScan mode ( BaseClear , Leiden , The Netherlands ) . Trace data were analyzed using the GeneMapper software package ( Applied Biosystems ) . Signals below the threshold value for noise cutoff in GeneMapper ( peak area <200 ) were assigned the threshold value for noise cut off . Subsequently , results from target genes were calculated relative to the average signal of GAPDH and assigned the threshold value if below 200 . HLA-E restricted T-cell clones ( 1x10e6 c/w ) were cultured in 24 well plates in the absence or presence of adherent macrophages ( 2 . 5x10e5 c/w ) . Macrophages were generated from CD14+ monocytes that were isolated from HLA-A2 negative buffycoat PBMCs with CD14 MACS beads ( Miltenyi ) and differentiated for 6 days in the presence of 50 ng/ml M-CSF ( R&D systems ) . Cultures were incubated with medium , specific or control peptide ( 10 μg/ml ) ( macrophages present ) , or macrophages infected with live BCG from fresh log culture ( MOI = 10 ) , or T-cell expander beads ( Invitrogen ) for 24 hours ( no macrophages present ) . Supernatants were tested using the Human Cytokine , Chemokine and Immuno Cell Multiplex Assays and the Human CD8+ T-Cell Multiplex Assay ( Merck Millipore , Amsterdam , the Netherlands ) . Analyses were performed on a Luminex200 with Bioplex software ( Biorad , Veenendaal , the Netherlands ) . B-cells were isolated from PBMC’s of HLA-A2 negative donors using CD19 MACS beads ( Miltenyi ) . Purified CD19+ cells were plated in a 96 well round-bottom plate ( 5x10e4 c/w ) and cultured with 5x10e4 HLA-E restricted CD8+ T-cells in AIMV . Also recombinant human cytokine controls for activation of B-cells in the absence of T-cells were performed ( 10 ng/ml of IL-4 , IL-5 , IL-10 and or IL-13 was used ( all Peprotech ) ) as were the positive controls CpG ( 5 μg/mL CpG ODN2006 ( Life Technologies ) ) and αIgG/IgM complex ( 10 μg/ml , Jackson ImmunoResearch Laboratories inc . , Suffolk , UK ) . After 48 hours supernatants were harvested for determination of IL-6 levels by standard IL-6 ELISA ( Biosource/Invitrogen ) according to the manufacturer’s protocol . Cells were harvested and activation was assessed by flow cytometry with surface staining for CD80 , CD86 , CD40 , CD25 and HLA-DR . Cells were also stained for CD3 and CD8 to exclude T-cells from the analyses ( CD3-PE-TexasRed , CD19-PacificBlue ( both Invitrogen ) , CD8-HorizonV500 , CD80-PE-Cy7 , CD86-FITC , CD40-APC , CD20-APC-H7 and HLA-DR-PE-Cy5 ( all BD biosciences ) . To investigate the specificity of B-cell activation by our HLA-E restricted Mtb specific T-cell clones we performed a similar co-culture assay using well-characterized classical CD4+ T-cell clones generated previously . Mycobacterium specific CD4+ T-cell clones: R2F10 ( reactive with Mtb hsp 65 , HLA-DR2 restricted ) produces Th1/2 cytokines , R2G7 ( reactive with Mtb hsp10 , HLA-DR2 restricted ) Th17 cytokine profile [31]; Rp15 1-1 ( reactive with Mtb hsp65 , HLA-DR3 restricted ) [64] Th1 cytokine profile; D2C4 ( reactive with Mtb 14–22 kD fraction , HLA-DQ restricted ) Th1 cytokine profile [65] . In addition to mycobacterium specific clones , we also included a Th1 CD4+ T-cell clone HA1 . 7 which recognizes influenza hemagglutinin in HLA-DR1 [66] and a cytolytic CD8+ T-cell clone HY21 . 17 recognizing the male HY antigen presented by HLA-A2 [23 , 67] . To address which T-cell cytokines were critical for B-cell activation , antibodies or isotype controls to block secreted cytokines were added to the B cell cultures at 20 μg/ml 2 hours prior to addition of the T-cell clones ( αIL-5 ( clone 14611 ) and αIL-13 ( clone 31606 ) ( R&D Systems ) , αIL-4 ( clone MP4-25D2 ) , αIL-10 ( cloneJES3-19F1 ) , isotype controls rat/ mouse IgG1 and rat IgG2a ( BD Biosciences and R&D systems ) ) and after 48 hours supernatants were collected for determination of IL-6 production and cells were stained as described above . Peripheral blood was obtained from 19 adults with TB disease ( 12 men , 7 women , age range 42–61 years ) from the Dipartimento di Medicina Clinica e delle Patologie Emergenti , University Hospital , Palermo . TB-infected patients had clinical and radiological findings consistent with active pulmonary TB ( American Thoracic Society , 2000 ) . Diagnosis was confirmed by bacteriological isolation of M . tuberculosis in 11 patients . Other patients were classified as having highly probable pulmonary TB on the basis of clinical and radiological features that were highly suggestive of TB and unlikely to be caused by any other disease; the decision was made by the attending physician to initiate anti-TB chemotherapy , which resulted in an appropriate response to therapy . All patients were treated in accordance with Italian guidelines and received therapy for 6 months . Treatment was successful in all participants all of whom completed the full course of anti-TB chemotherapy , as shown by the absence of any clinical or radiographic evidence of recurrent disease and sterile mycobacterial cultures . Peripheral blood was collected before chemotherapy . None of the TB patients had been vaccinated with Bacillus Calmette-Guerin ( BCG ) , or was being treated with steroid or other immunosuppressive or anti-tubercular drugs at the time of their first sampling . Three patients had evidence of HIV infection . Tuberculin ( PPD ) skin tests were considered positive when the induration diameter was larger than 10 mm at 72 hrs since injection of 5 U of PPD ( Statens Seruminstitut , Copenhagen , Denmark ) . The study was approved by the Ethical Committee of the University Hospital , Palermo , where the patients were recruited . The study was performed in accordance to the principles of the Helsinki declaration and those of the “Good Clinical Practices” , and all individuals gave written informed consent to participate . Tetramer staining was carried out as described in detail previously [68 , 69] . PBMC ( 106/mL ) were incubated in U-bottom 96-well plates , washed twice in phosphate buffered saline ( PBS ) containing 1% fetal calf serum ( FCS , Sigma ) and stained for 30 min at 4°C with PE-labelled tetramers ( 5μL each ) prepared as previously described [68–70] , washed and subsequently stained with FITC-labelled anti-CD8 mAb ( RPA-TB , BD Biosciences ) and analyzed by flow cytometry on a FACSCanto . Data were analyzed with the use of FACSDiva ( BD Biosciences ) . Viable lymphocytes were gated by forward and side scatter and the analysis was performed on 100 , 000 acquired CD8 events for each sample . A cut-off of 0 . 01% was used as described previously [71]; values below this were set to zero . Cytokine production following peptide stimulation was analysed by intracellular staining and flow cytometry [71] . PBMCs ( 106/mL ) were stimulated with peptides for 16 hours , the last 12 hours in the presence of monensin at 37°C in 5% CO2 . The cells were harvested , washed and stained with anti-CD8 mAb ( RPA-TB , BD Biosciences ) in incubation buffer ( PBS-1% FCS-0 . 1% Na azide ) for 30 min at 4°C . The cells were washed twice in PBS-1% FCS and fixed with PBS-4% paraformaldehyde overnight at 4°C . Fixation was followed by permeabilization with PBS-1% FCS-0 . 3% saponin-0 . 1% Na azide for 15 min at 4°C . Staining of intracellular cytokines was performed by incubation of fixed permeabilized cells with anti-IFN-γ ( 25723 . 11 , BD Biosciences ) , anti-IL-2 ( MQ1-17H12 , BD Biosciences ) , anti-TNF-α ( MAb11 , BD Biosciences ) , anti-IL10 ( BT-10 , eBioscience ) , anti-IL-17A ( eBio64DEC17 , eBioscience ) , anti-IL-4 ( BD Biosciences , 3010 . 211 ) , anti-IL13 ( Biolegend , JES10-5A2 ) mAbs or isotypematched control mAbs , all from BD Bioscience . Cells were acquired and analyzed by FACS as described above . Analysis was performed on a minimum of 100 , 000 acquired CD8+ events for each sample . Negative controls were background staining obtained with medium , in the absence of any stimulant . Cut-off values for a positive response were predetermined to be in excess of 0 . 01% responsive cells . Results below this value were considered negative and set to zero . Groups were compared using Mann-Whitney U test and p<0 . 05 was considered significant .
|
Pathogens like Mycobacterium tuberculosis ( Mtb ) are recognized by human T-cells following their presentation in HLA molecules . HLA class I molecules can be divided into two types , classical as well as non-classical HLA molecules . Here we studied the non-classical HLA family member , HLA-E , which displays only minimal genetic variation between individuals and is relative resistant to down modulation by HIV infection . We have characterized the T-cells that recognize Mtb in the context of HLA-E in detail and found that these human CD8+ T-cells had unexpected , unorthodox properties: in contrast to most classical CD8+ T-cells , the T-cells activated by HLA-E uniquely produced Th2 ( IL-4 , IL-5 , IL-13 ) instead of the usual Th1 cytokines , and were able to activate B-cells and induced cytokine production by these B-cells . Moreover , these HLA-E restricted CD8+ T-cells inhibited Mtb growth inside cells , an important property to contribute to resolution of the infection . Thus these T-cells represent a new player in the human immune response to infection , and add B-cell activation to the key pathways following infection with Mtb .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Human CD8+ T-cells Recognizing Peptides from Mycobacterium tuberculosis (Mtb) Presented by HLA-E Have an Unorthodox Th2-like, Multifunctional, Mtb Inhibitory Phenotype and Represent a Novel Human T-cell Subset
|
Zika virus ( ZIKV ) , was widely reported in Latin America and has been associated with neuropathologies , as microcephaly , but only few seroprevalence studies have been published to date . Our objective was to determine the seroprevalence amongst Bolivian blood donors and estimate the future potential circulation of the virus . A ZIKV seroprevalence study was conducted between December 2016 and April 2017 in 814 asymptomatic Bolivian volunteer blood donors residing in various eco-environments corresponding to contrasting entomological activities . It was based on detection of IgG to ZIKV using NS1 ELISA screening , followed by a seroneutralisation test in case of positive or equivocal ELISA result . Analysis revealed that ZIKV circulation occurred in tropical areas ( Beni: 39%; Santa Cruz de la Sierra: 21 . 5% ) but not in highlands ( ~0% in Cochabamba , La Paz , Tarija ) . It was modulated by Aedes aegypti activity and the virus spread was not limited by previous immunity to dengue . Cases were geo-localised in a wide range of urban areas in Santa Cruz and Trinidad . No differences in seroprevalence related to gender or age-groups could be identified . It is concluded that ZIKV has been intensely circulating in the Beni region and has still a significant potential for propagating in the area of Santa Cruz .
Zika virus ( ZIKV ) is an arthropod-borne Flavivirus transmitted to humans mainly by Aedes mosquitoes . It has been responsible over the last decade for outbreaks in Pacific Islands [1] ( Yap Island , 2007 [2]; French Polynesia , 2013 [3]; New Caledonia , Cook Islands and Easter Island , 2014 [4]; Vanuatu , Solomon Islands , Samoa and Fiji , 2015 [5] ) , in Latin America ( from late 2013 [6] or early 2014 [7] in Brazil , then in a large number of other countries [8] ) , and in the Caribbean region ( e . g . , 2014 in Haiti [9] , 2015 in Martinique Island [10] ) . The first autochthonous case reported in Bolivia was in January 2016 [11] . ZIKV is of African origin , but an Asian lineage emerged presumably in the first part of the XIXth century [6] and the viruses that spread in the Pacific and the Americas are descendants of this lineage . In addition to the classical picture of large arboviral outbreaks , significant public health burden was endured in Polynesia and America when severe and formerly undescribed foetal and neurological complications of the disease [12] as well as non-vectored routes of transmission [13] were reported . Our capacity to estimate the future spread of ZIKV disease in South America significantly depends on our knowledge of the immune status against ZIKV in the populations exposed to the potential vectors ( mostly Aedes aegypti mosquitoes ) . Few ZIKV seroprevalence results in the region have been made available yet , most probably due to the technical difficulty to distinguish antibodies to ZIKV from cross-reacting antibodies to the other flaviviruses , in particular dengue virus . Specific detection of antibodies to ZIKV can be improved when using demanding seroneutralisation methods for the primary detection of antibodies , or for confirmation of a more convenient screening assay such as an ELISA test . In this context , we investigated the ZIKV serological status of blood donors from different regions of Bolivia and analysed results with reference to immunity of the same populations to dengue ( DENV ) and chikungunya virus ( CHIKV ) , two arboviruses known to be transmitted by the same vector locally . DENV has been reported in Bolivia since 1931 . In 1948 , Bolivia was declared Ae . aegypti free , but the vector reappeared in the 1980s . Since then , DENV-1 , DENV-2 , DENV-3 , lately DENV-4 and the co-circulation of serotypes have been reported , being massively endemic in the tropical regions of Bolivia [14 , 15] . CHIKV , a member of the Alphavirus genus arrived in the Caribbean in late 2013 and then spread through Latin America the following years , causing explosive outbreaks in humans[16] .
We conducted a study with the help of five Bolivian regional blood banks , representing a variety of eco-environments ( Santa Cruz de la Sierra and Beni have tropical climate; Cochabamba , Tarija and La Paz have colder subtropical highland climates ) : 814 volunteer blood donors from Santa Cruz de la Sierra ( n = 200 ) , La Paz ( n = 161 ) , Cochabamba ( n = 152 ) , Tarija ( n = 196 ) , and Beni ( n = 105 ) provided before blood donation their oral consent for detection of IgG to ZIKV . All donors accepting to participate in the study and providing consent were considered eligible . No specific sampling was performed and leftovers of blood samples collected and stored at -20°C after completion of laboratory analyses were used . The contribution of each site was evaluated locally according to the actual possibilities to recruit donors , process samples and provide the epidemiological data requested ( see below ) . The sample size was in agreement with previous epidemiological studies of arbovirus seroprevalence in other countries and expected to provide a reliable picture of the global epidemiological situation [17 , 10] . Sampling was performed in December 2016 in the Beni region , then from March to April 2017 in the other sites , according to local logistical possibilities . Blood samples and personal data ( date of donation , sex , age , birthplace , living place , occupation and neighbourhood ) were irreversibly anonymised . Adults blood donors approved to participate in the study by providing oral consent during the face-to-face questioning before blood gift . This procedure was considered the most suitable by local blood banks . The sampling and analysis protocol was approved by the ethics committee of the Medical College of Santa Cruz . Samples were tested for the presence of IgG to ZIKV as previously described [10 , 17] . In brief , alike Netto and collaborators[18] we performed an initial screening with a recombinant NS1-based ELISA test ( Euroimmun , Lübeck , Germany ) [19 , 20] which is the only test certified for serological diagnostics of ZIKV by the responsible Brazilian authority ANVISA ( Agência Nacional de Vigilância Sanitária ) [18] and a subsequent Virus Neutralisation Test ( VNT ) for samples with a positive or equivocal ELISA result ( ratio ≥0 . 8 ) . VNT was performed in a 96-well format based on cytopathic effect ( CPE ) , using ZIKV strain H/PF/2013[21] , Vero ATCC cells monolayers and serum dilutions from 1:10 to 1:320 . All samples were tested in duplicate with positive and negative serum controls . Sera with titre ≥1:40 were considered positive , according to the recommendations of the French National Reference Centre for Arboviruses . This testing strategy was previously demonstrated to provide specificity and sensitivity values above 98 . 5% in volunteer blood donors of Martinique Island tested before , during , and after the 2016 ZIKV outbreak[22] and with heavy exposure to dengue[23] . To estimate the serological immune background against dengue and chikungunya ( which in Bolivia are also transmitted by Aedes aegypti ) , we randomly selected approximately half of the samples on each site ( Beni , n = 60; Santa Cruz de la Sierra , n = 108; Tarija , n = 111; La Paz , n = 93; Cochabamba , n = 77; i . e . , 449 in total ) . They were tested for the presence of IgG to DENV and CHIKV ( Euroimmun dengue ELISA IgG and chikungunya ELISA IgG assays ) according to the manufacturer's recommendations . Donors were assigned for analysis to 3 age-groups ( 18–30 years-old; 31-40yo; 41-60yo ) and 6 ZIKV ELISA ratio groups ( <0 . 8; 0 . 8–1 . 09; 1 . 1–2 . 49; 2 . 5–3 . 99; 4 . 0–5 . 49; ≥5 . 5 ) . Statistical analyses were performed using IBM-SPSS Statistics v 24 . 0 . 0 . 0 software . Statistical association between ZIKV seropositivity and age-group or gender was evaluated , as well as relationship between ZIKV seropositivity and DENV or CHIKV seropositivity ( Chi square test , significant threshold: p = 0 . 05 ) .
Serological results in the different sites for ZIKV are shown in Fig 1 . ZIKV has been circulating in the two regions with a typical tropical climate ( Beni and Santa Cruz de la Sierra ) , but not in highlands in which the entomological activity is limited . In the tropical regions , the usual period of circulation of Aedes borne viruses ranges from November to April . The high seroprevalence rate in Beni ( 39% after VNT , ( 95% CI [30%–48%] ) ) is in agreement with the report of an outbreak of febrile cases with rash locally , with laboratory PCR confirmed cases before our study ( performed in December 2016 ) . Since cases were reported in Beni until the end of the rainy season , it is likely that the final seroprevalence rate is even higher nowadays in this region . The significant rate in Santa Cruz ( 21 . 5% after VNT , ( 95% CI [16%-27%] ) ) is in line with the report of PCR confirmed cases locally before the collection of samples ( in March and April 2017 ) . Detection of cases benefitted from the local implementation of the Cenetrop National Reference Laboratory , but no clear epidemic pattern was reported . Since the study was performed locally at the end of the rainy season , it is most probable that the seroprevalence rate remained locally at a lower level than in Beni . Of note , the proportion of ELISA positives confirmed by VNT was much higher in Beni than in Santa Cruz ( 63 . 1% vs 35 . 2% ) , possibly reflecting the frequent and intense immune stimulation against ZIKV in the Beni population . When examining the relationship between ZIKV ELISA and VNT results , it appears , as previously observed [10 , 17] , that the rate of VNT confirmation increases with the value of the ELISA ratio , reaching 65% for an ELISA ratio ≥4 and 95% for an ELISA ratio ≥5 . 5 ( S1 Table ) . Table 1 shows the distribution of seropositives per site with the background of serological results for dengue and chikungunya , which circulation in Bolivia has been widely documented . Immunity to dengue virus is the rule in the tropical regions ( >90% of seropositives in Beni and Santa Cruz ) , but also significant in the region of Tarija ( 44 . 1% ) . It is more limited in Cochabamba and La Paz ( ca ~10% ) . For chikungunya , approximately half of the population is seropositive in the tropical regions and less than 10% in highlands . The differences observed with dengue most probably reflect the lower number of epidemic waves of chikungunya that hit the country . Dengue was first reported in Bolivia in 1931 and since then has been circulating intensely with major epidemics in the tropical regions[15] and , over time , multiple imported cases and possibly transient local transmission in highlands ( in particular the Tarija region ) . By contrast , autochthonous transmission of CHIKV was first reported in Bolivia as recently as 2015 following the introduction of the virus in the Americas[24] , limiting herd immunity outside the areas of intense epidemic transmission . Altogether , the spreading pattern of ZIKV follows that of other Aedes aegypti-borne viruses in Bolivia , in particular that of the recently introduced CHIKV . S2 Table details the distribution of seropositives for ZIKV , DENV and CHIKV according to sex and age groups . Differences are minimal between groups ( none reaches a significance threshold of 0 . 05 ) , suggesting that age and sex do not significantly impact exposure to these arboviral diseases in the population investigated . Importantly , there is in the population studied a strong relationship between seropositivity to ZIKV and to DENV ( p<10−12 ) , but also to CHIKV ( p<10−15 ) , obviously pointing to exposure to a common risk factor: the bite of the Aedes aegypti vector of the three diseases . This study has classical limitations linked to the sampling procedure in blood donors ( in particular data regarding individuals under the age of 15 years old and in pregnant women could not be obtained ) . However , many previous studies on arboviruses including dengue virus[23] chikungunya virus[25] , Zika virus [10] have suggested that blood donors constitute a valuable population to identify the major epidemiological trends that underlie exposure to arbovirus transmission and spread . Amongst great advantages of studying blood donor's populations , one can mention the logistical capability to perform rapidly multisite studies in the absence of robust local research infrastructure , the absence of the need to perform specific sampling , and the access to comparable populations in multiple sites that allows comparison of prevalence values . We conclude that this seroprevalence study confirms the circulation of Zika virus in Bolivia . Despite previous reports of the presence of A . aegypti in all 5 departments investigated [14 , 26 , 27 , 28 , 29] , the potential for epidemic spread is deeply modulated by the variable entomological activity in the different locations , in relation with different eco-environments ( and in particular different altitudes ) and as reflected by contrasted exposure to dengue or chikungunya . In the tropical areas of Santa Cruz and Beni , cases were identified from the city centres to outlying district , reflecting the wide distribution of A . aegypti . According to previous information relating to the circulation of dengue in Bolivia and corroborated by our present seroprevalence data , the spread of Zika virus was not limited by previous herd immunity to dengue virus . However , it remains possible that prior immunity to DENV modified the epidemiological pattern of the virus global spread . The same methodological protocol has been used previously to estimate ZIKV seroprevalence in Martinique Island [10] ( Caribbean region ) and Cameroon[17] ( Central Africa ) . Clearly , the transmission pattern in Bolivia is very different from the ( peri- ) sylvatic transmission reported in Cameroon , and more closely related to the urban transmission by the ( peri- ) domestic A . aegypti in Martinique . With reference to the Martinique outbreak and seroprevalence study , a minimum seroprevalence rate around 50% seems to be required to provide herd immunity that can stop ZIKV circulation . Accordingly , the ~21% rate observed in Santa Cruz ( in the last phase of the arbovirus circulation period ) would be insufficient to give protective herd immunity in the presence of abundant potential vectors and intense entomological activity , and with sustained circulation and potential reintroduction of the virus in Latin America . It is therefore expected that ZIKV circulation should be limited in the near future in the Beni Region ( data were collected in December and the virus could circulate for at least four additional months locally , therefore the final seroprevalence rate may be even higher ) , but , in contrast , ecological and epidemiological conditions are favourable for further circulation of the virus in Santa Cruz , with its consequent complications especially in pregnant women .
|
Zika virus ( ZIKV ) is a virus of African origin , transmitted by Aedes mosquitoes , and related to dengue and yellow fever virus . It was originally believed to be responsible for a mild febrile illness in Africa and South-east Asia . However , in recent years , ZIKV has been responsible for outbreaks in the Pacific Islands before massively spreading in Latin America and the Caribbean . On this occasion , ZIKV has unexpectedly been associated with non-vector transmission ( i . e . , sexual and mother-to-foetus transmission ) and with severe complications such as foetal abnormalities ( e . g . microcephaly ) and Guillain-Barré syndromes . Little is known about the actual proportion of the populations infected by ZIKV in Latin America . Here , we report a seroprevalence data in this region , after studying 814 asymptomatic Bolivian volunteer blood donors residing in various eco-environments corresponding to contrasting entomological activities . We conclude that ZIKV has been circulating in Bolivian tropical areas but not in highlands , and that the epidemic has not been limited by previous immunity against dengue . Specific attention should be paid to the region of Santa Cruz , where the seroprevalence is still limited , but the density of Aedes aegypti populations makes plausible further spreading of the disease .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
] |
[
"dengue",
"virus",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"pathogens",
"immunology",
"geographical",
"locations",
"microbiology",
"animals",
"alphaviruses",
"health",
"care",
"viruses",
"chikungunya",
"virus",
"rna",
"viruses",
"immunologic",
"techniques",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"south",
"america",
"aedes",
"aegypti",
"medical",
"microbiology",
"microbial",
"pathogens",
"immunoassays",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"eukaryota",
"flaviviruses",
"immunity",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"blood",
"donors",
"species",
"interactions",
"bolivia",
"organisms",
"zika",
"virus"
] |
2018
|
Zika virus epidemiology in Bolivia: A seroprevalence study in volunteer blood donors
|
The recent explosion of genomic data has underscored the need for interpretable and comprehensive analyses that can capture complex phylogenetic relationships within and across species . Recombination , reassortment and horizontal gene transfer constitute examples of pervasive biological phenomena that cannot be captured by tree-like representations . Starting from hundreds of genomes , we are interested in the reconstruction of potential evolutionary histories leading to the observed data . Ancestral recombination graphs represent potential histories that explicitly accommodate recombination and mutation events across orthologous genomes . However , they are computationally costly to reconstruct , usually being infeasible for more than few tens of genomes . Recently , Topological Data Analysis ( TDA ) methods have been proposed as robust and scalable methods that can capture the genetic scale and frequency of recombination . We build upon previous TDA developments for detecting and quantifying recombination , and present a novel framework that can be applied to hundreds of genomes and can be interpreted in terms of minimal histories of mutation and recombination events , quantifying the scales and identifying the genomic locations of recombinations . We implement this framework in a software package , called TARGet , and apply it to several examples , including small migration between different populations , human recombination , and horizontal evolution in finches inhabiting the Galápagos Islands .
Since the publication of the first draft of the human genome [1 , 2] , there has been an explosion in genomic data . The genomes of thousands of different human individuals have been sequenced [3] , several hundreds of eukaryotic genomes have been characterized , and new viral , bacterial and archaeal species are being sequenced on an almost daily basis [4 , 5] . Darwin provided a historical dimension to the taxonomical enterprise , proposing that closely related species in the hierarchical taxonomy share ancestors . Since then , tree-like structures have been proposed to represent the evolutionary/historical relationship between organisms . In the last few years , however , the richer and more comprehensive genomic characterization of many organisms have underscored the need of representations that are not strictly tree-like . Phenomena such as horizontal gene transfer in bacteria [6] , the ability of viruses to borrow and lend genes across species , and hybridization in metazoa ( in plants , in particular [7 , 8] ) are exposing some of the limitations imposed by tree-like phylogenetic structures . The definition of species itself becomes cumbersome in bacteria and viruses [9] . Within many species , including humans , genetic recombination is so pervasive that tree-like representations are useless . It is then natural to wonder what other frameworks could be used to capture phylogenetic relationships without losing the interpretability and simplicity of trees [10–12] . Of particular interest are representations that reduce to trees when evolution is tree-like; that capture genetic relations between ancestors , and identify genomic regions originating from different ancestral lineages; and , more generally , that allow for an interpretation of the observed data in terms of a chronological sequence of events . Several such frameworks have been proposed in the last two decades . The study of phylogenetic networks has been an area particularly active [13–15] . Phylogenetic networks provide representations that extend trees to graphs ( networks ) , generating loops when the data does not fit into a tree . Some of those methods can easily be applied to more than one hundred genomes [16–21] providing the opportunity for large-scale representations . However , the biological interpretation of these representations is limited , as loops represent inconsistencies with trees , but it is unclear how these inconsistencies arose historically , what genomic regions were involved , or how frequently an exchange happened . Other types of representations , sometimes named explicit networks [13 , 22] , do aim to provide a historical account in terms of a chronology of events . Ancestral recombination graphs ( ARGs ) provide potential explanations of the observed data in terms of a progression of recombination and mutation events . As in trees , mutations are represented as events along the branches . Recombinations , however , appear as the fusion of two parental branches into one offspring branch . ARGs provide simple histories that can be used in association mapping [23–25] , SNP genotyping [26] or inference of the frequency and scale of recombination [27] . However , these applications are hindered by the computational infeasibility of constructing ARGs that explain hundreds of sequences . The construction of minimal ARGs , containing the minimum number of recombination events required to explain the sample in absence of convergent evolution and back-mutation , is an NP-hard problem [28–30] . Several approximations have been developed in the last few years , including galled trees [31 , 32] , branch and bound [33] , heuristic [23] and sequentially Markov coalescent approaches [34] . Recently , a new framework to study genomic relationships has been proposed [35–37] , based on topological data analysis [38–40] . Topology is the area of mathematics that aims to characterize properties of spaces up to continuous deformations , for instance the number of disconnected components , loops and holes of a space . TDA extends the concepts and tools of topology to finite metric spaces , that is , finite sets of points and distances between them . Taking the premise that a set of points has been sampled from an unknown underlying space , TDA attempts to infer the topological features of the space ( Fig 1A ) . Stability results [35 , 41 , 42] guarantee that small fluctuations in the data only create small changes in the inferred topological features , providing robust characterizations of the data . In a TDA framework , genomes are characterized by points in a high dimensional space where pairwise distances are genetic distances between sequences . Assuming that each genomic site mutates at most once across the evolutionary history of the sample , the genetic distance between two genomes can only increase with the acquisition of novel mutations . The only way of “closing” a loop ( a close path ) in this space is therefore by means of a recombination event [35] . Hence , an approach to studying recombination in the sample of genetic sequences is to study the loops that those sequences generate when represented in the above way . A valuable attribute of TDA methods is that they are informative about the scale or size of the inferred topological features . Given a finite set of data points , there is an infinite number of spaces that are compatible with the points . TDA structures this spectrum of possibilities by introducing a notion of scale ( Fig 1B ) : at a given scale ϵ , two points are connected in the underlying space if their distance is smaller than ϵ . Topological features compatible with the data can be then summarized in terms of sets of intervals , named barcodes [43] ( Fig 1C ) . Each interval in a barcode represents the range of scales across which a particular topological feature ( e . g . a loop ) is present in the inferred topological space . In the genomic context introduced above , barcodes of loops summarize the frequency and scale ( mutational distance between recombining sequences ) of recombination events , and provide a basic structure on which statistics of genomic exchange can be built [37] . TDA methods are particularly well suited for large datasets . In the context of molecular phylogenetics and evolution , they have been applied to the study of viral recombination and reassortment [35] , bacterial species [36] and point estimators in population genetics [37] . However , these implementations of TDA have limitations , as they are not tailored for the biological problem they try to address . Specifically , traditional TDA methods only use information about genetic distances between sequences , and so they discard the full structure of segregating characters , missing numerous recombination events that are required to explain the data . Relatedly , it is unclear which specific evolutionary histories explaining the data TDA informs about , and what is the precise relation between barcodes and these histories . Here we address these two important aspects , improving on the scalable capabilities of TDA to extract robust information on the possible evolutionary histories of a sample of genetic sequences . In particular , we show that by systematically sampling subsets of segregating sites and performing TDA , we are able to identify most of the necessary recombination events identified by bound methods [33 , 44 , 45] , providing a significant improvement of past methods [35–37] in terms of interpretation and sensitivity . Moreover , we introduce a novel type of graph ( topological ARG or tARG ) , closely related to minimal ARGs , that captures ensembles of minimal recombination histories; and we show that TDA informs about the topological features and genetic scales of these graphs . Like minimal ARGs [22 , 23] , tARGs can be considered as explicit , parsimonious , interpretable phylogenetic representations . The main advantage of tARGs and barcodes versus minimal ARGs is , however , the possibility of obtaining such phylogenetic information in polynomial time , which allows us to deal with hundreds of sequences . We have implemented this method in a software , called TARGet , and have illustrated it with several examples , including small migration between diverging populations , human recombination , and horizontal evolution of finches inhabiting the Galápagos archipelago . The software , instructions and example files used in the manuscript can be obtained from https://github . com/RabadanLab/TARGet .
An ARG is an explicit phylogenetic network representing a possible evolutionary history of a sample of genetic sequences , where only mutation and recombination events are present and convergent evolution is not considered and so never occurs [22 , 46 , 47] . ARGs are very useful constructs in population genetics and phylogenetics . However , the problem of building a minimal ARG from a set of genetic sequences is known to be NP-hard [28–30] . The use of ARGs has therefore been traditionally limited to small samples , consisting of a handful of sequences . In this section , we introduce a particular class of minimal ARGs and a set of related graphs . Then , using computational algebraic topology , in the next section we show that it is possible to extract , in polynomial time , phylogenetic information from this class of minimal ARGs , without having to explicitly construct them . Thus , by restricting to this specific class of graphs , we are able to extend the realm of ARGs to large samples of sequences . To be specific , we consider a sample S consisting of n distinct genetic sequences with m binary segregating characters . The latter can be single nucleotide polymorphisms ( SNPs ) , indels , gene duplications or any other genetic trait that takes one of two possible states , 0 or 1 , in each sequence . An ARG is then formally defined as a directed acyclic graph N with n leaf nodes and a unique root node , where every node other than the root has in-degree one ( tree node ) or two ( recombination node ) , every segregating character labels a unique edge in N ( infinite sites assumption ) , and every sequence in S labels a unique leaf in N . Moreover , each node in N is labelled by a m-length binary sequence , such that the sequence labelling a tree node differs from the sequence of the parent node only at the character labelling the edge that connects the two nodes; and the sequence labelling a recombination node is a combination of the sequences labelling the two parent nodes . Single-crossover recombinant sequences are formed by taking the first k sites from the sequence of one of the parent nodes ( prefix ) and appending the last m − k sites from the sequence of the other parent node ( suffix ) , for k ∈ [1 , m − 1] . There is an infinite number of ARGs that can explain a given sample S [22] . A stochastic model , such as the coalescent model with recombination [46 , 48] , would assign probabilities to each possible ARG . Here , however , we adopt a parsimony approach and consider ARGs that are minimal ( in a sense defined below ) , without assuming an underlying probabilistic model . Such a model-independent approach has proven useful in summarizing genetic sequences into evolutionary histories where all events are required . Specifically , we consider ARGs that contain exactly the minimum number Rmin of single-crossover recombinations required to explain the sample , and that minimize the function D ( N ) = ∑ r = 0 R min d r ( 1 ) where the sum runs over all recombination events in N , and dr is the Hamming distance between the two parental sequences involved in the r-th recombination . This is a more restricted definition of minimal ARG than the one that usually appears in population genetics literature [22] , where the condition on D ( N ) is generally not required . We use the term ultra-minimal ARG to refer to this restricted type of minimal ARG . Ultra-minimal ARGs are thus minimal ARGs where recombination events involve parental sequences that are as genetically close as possible . They introduce a higher level of parsimony than minimal ARGs , being informative not only about the minimum number of recombination events , but also about the minimum genetic distance between the recombining sequences that took part in those events . By construction , an ultra-minimal ARG explaining any given sample always exists . Examples are shown in Figs 2 and 3 . A minimal ARG can be condensed by collapsing all unlabelled edges , so that the resulting graph can be embedded into an m-dimensional hypercube and its diagonals ( that is , the line segments joining non-consecutive vertices ) ( Fig 2 ) . The number of edges and vertices of such a condensed representation is m + 2Rmin and m + Rmin + 1 , respectively , whereas the number of independent loops is Rmin , where a loop is said to be independent if it cannot be embedded in the union of other loops . In this representation , the distance between two nodes is defined as the number of edges in the shortest path connecting the nodes , and is equal to the Hamming distance between the corresponding sequences . Given a sample S of genetic sequences , we would like to obtain information about the ultra-minimal ARGs that explain S , without explicitly constructing them . To that end , we consider the undirected graph G = ( V , E ) , with vertices V and edges E = E1 ∪ … ∪ El , that results from the union of all condensed ultra-minimal ARGs G i = ( V , E i ) explaining S and having the same set of vertices V ( Fig 4 ) . We call this construction topological ARG ( tARG ) . A tARG therefore summarizes the collection of most parsimonious histories associated to a sample of genetic sequences . However , unlike minimal ARGs , tARGs are completely determined by their vertices . By considering tARGs instead of minimal ARGs , we are able to reduce an NP-hard problem into a much simpler ( but still very informative ) topological problem , as we describe in next section . Topological data analysis has emerged during the last decade as a branch of applied topology that attempts to infer topological features of spaces ( such as the number of loops and holes ) from sets of sampled points [38] . The topological features of a space are preserved under continuous deformations of the space and can be arranged in mathematical structures called homology groups [49] . We refer the reader to refs . [49 , 50] for formal definitions and basic introductions to algebraic topology . In brief , the nth homology group of a space is an algebraic structure that encompasses all ( n + 1 ) -dimensional holes of the space . Of special interest to us is the first homology group , whose elements correspond to loops . Homology groups can be computed by replacing the original space with a simpler one , known as simplicial complex , which has the same topological features as the original space but consists of a finite set of elements ( Fig 1B ) . A simplicial complex is a generalization of a network that , in addition to nodes and vertices , includes higher dimensional elements like triangles and tetrahedra . Simplicial complexes are powerful because they allow the implementation of algebraic operations to extract the topological features of the space . When only a finite set of points of the space is given , there is still a well-defined notion of homology groups , known as persistent homology [39 , 40] , which capture the topological features of the underlying space . At each value of a scale parameter ϵ , a simplicial complex ( known as Vietoris-Rips complex ) can be constructed by considering the intersections of balls of radius ϵ centred at the sampled points ( Fig 1B ) . Points are joined if their corresponding balls intersect . This process produces a sequence of simplicial complexes parametrized by ϵ , from which persistent homology can be computed using available algorithms [39 , 40] . Remarkably , the computation time of persistent homology is polynomial in the number of points [39 , 40] . Persistent homology can be represented using barcodes [43] . These are graphical representations where each element of persistent homology is represented by a segment spanning the interval [ϵb , ϵd] , where ϵb and ϵd are the values of the parameter ϵ at which the corresponding feature is respectively formed and destroyed in the sequence of simplicial complexes ( Fig 1C ) . Thus , each segment in a barcode represents a topological feature inferred from the data , and the position and length of the segment are informative of the size of the topological feature . The values ϵb and ϵd are referred as birth and death time of the topological feature , respectively . In the current context , we exploit the use of persistent homology to infer topological features of an unknown tARG , given a set of sampled nodes ( Fig 5 ) . The use of persistent homology to detect the presence of recombination in genetic samples was proposed in [35] . However , the relation between persistent homology and explicit evolutionary histories incorporating recombination events was not studied . Our aim is inferring information about the loops of the tARG , as they correspond to recombination events present in the collection of most parsimonious histories explaining the sample . To that end , we consider the Hamming distance matrix of the sample and compute persistent homology using the algorithm developed in ref . [39 , 40] . Since computing the distance matrix and persistent homology requires respectively O ( n 2 m ) and O ( n 3 ) operations [39 , 40] , the running time grows at most cubically with the number of genetic sequences . An advantage of using persistent homology instead of just counting loops in a nearest neighbour graph is that we also obtain valuable information about the genetic distances between recombining sequences . The barcode that results from this computation contains information about the number and size of the loops in the tARG underlying the sample ( Fig 5 ) . Each segment in the barcode represents a loop in the tARG , and therefore a recombination event in an ultra-minimal ARG explaining the sample . The position of each segment provides information about the genetic scales involved in the corresponding recombination event . Specifically , 2ϵd sets an upper bound to the mutational distance between the two recombining sequences , since all pairwise distances between nodes in the loop are smaller than 2ϵd . The number of segments in the barcode ( namely , the dimension of the first persistent homology group ) or persistent first Betti number , b1 , is hence a lower bound of the number of recombination events in the tARG , R ¯ min . Note that , since a tARG is the union of multiple minimal histories , R ¯ min can be larger than Rmin . In particular , R ¯ min > R min when there are three characters for which all eight possible allele combinations appear in the sample . In general , this can only happen at very large recombination rates . The sensitivity of persistent homology to detect recombination decreases as the number m of segregating characters increases . Indeed , in that case the dimensionality of the ambient space is larger and the sample becomes sparser . For this reason , b1 is in general a loose lower bound of R ¯ min . To address a similar problem , Myers and Griffiths introduced the idea of combining the local bounds that result from partitioning the sequence , building a more stringent global bound [45] . In this way , information about the ordering of characters is incorporated and the location of recombination breakpoints is constrained in the sequence . This general idea was applied in [45] to the haplotype bound , n − m − 1 ≤ Rmin , to built a stronger lower bound of Rmin , denoted RMG . A similar idea can be applied in the context of barcodes to build a barcode ensemble , given by the disjoint union of the persistent first-homology barcodes of a set of optimally chosen , non-overlapping intervals within the sequence alignment ( Fig 6A ) . Given a partition of a genetic sequence , the barcode associated to each interval captures information about recombination events with breakpoint in that interval . Due to the curse of dimensionality mentioned in the previous paragraph , the union of the barcodes associated to two contiguous genomic intervals often captures more recombination events than the barcode associated to the union of the two genomic intervals . Therefore , by systematically exploring all possible partitions of the genetic sequence , it is possible to find a partition that maximizes the total number of bars in the barcodes . The solution is often not unique , as different partitions may lead to the same total number of bars . One may reduce this degeneration by considering additional criteria , such as also maximizing the total length of the bars ( so that they are more informative about genetic distances ) . The formal details of the barcode ensemble construction are presented in the Methods section . The barcode ensemble incorporates information about the full structure of characters in the sample , largely increasing the sensitivity of persistent homology to recombination and providing information on the location of the recombination breakpoints in the sequence . The number of bars in the barcode ensemble , b ¯ 1 , is an improved lower bound of R ¯ min , in the same way as RMG is an improved lower bound of Rmin: tARG → R ¯ min ≥ b ¯ 1 ≥ b 1 ↑ ( ultra ) minimal ARG → R min ≥ R MG ≥ n − m − 1 In biological data , b ¯ 1 and RMG are in general very close to each other ( Fig 6B ) , as tARGs with R ¯ min > R min occur very rarely . However , unlike RMG , barcode ensembles provide additional phylogenetic information , such as bounds on the mutational distances between recombining sequences ( note that birth and death times in barcode ensembles refer to local genetic distances , namely mutational distances across the genomic interval associated to the particular bar ) . These features put barcode ensembles at the very interesting interface between the fast , but phylogenetically limited , existing lower bounds to Rmin; and the slow , but phylogenetically rich methods for reconstructing minimal ARGs . We have implemented the computation of barcode ensembles in publicly available software , called TARGet . We consider five examples that illustrate how the formal developments presented in previous sections can be used to extract useful phylogenetic information from samples of genetic sequences . The first example is a simple toy model where an explicit minimal ARG can be easily constructed . It displays how the information contained in the barcode ensemble of the sample directly maps to features of ultra-minimal ARGs . The second example , based on simulated data of two sexually reproducing populations exchanging genetic material at low rate , shows the applicability of persistent homology to large datatsets , consisting of several hundreds of sequences . It also demonstrates the use of phylogenetic information contained in the barcode ensemble to distinguish among various biological settings with similar recombination rates . The third and fourth examples consist respectively of 250 and 100 kilobase regions in the HLA and MS32 loci of ∼ 100 humans , where several meiotic recombination hotspots localize . The fifth example consists of a 9 megabase scaffold in the genome of 112 Darwin’s finches [51] . These last three examples serve to illustrate the applicability of barcode ensembles to real datasets . The examples above illustrate the use and interpretation of barcode ensembles in molecular phylogenetics . As we have discussed , an important feature of topological approaches to phylogenetics is that they inform about most parsimonious evolutionary histories . Being model-independent approaches , they describe minimal sets of events required to explain a sample of sequences , without assuming any probabilistic model of evolution . In some situations , however , we are interested in estimating the parameters of a specific evolutionary model from the observed data ( e . g . the recombination rate in a coalescent model with recombination ) . To that end , barcode ensembles can be taken as summary statistics from which to build parameter estimators . For instance , in Fig 11A we show the dependence of b ¯ 1 on the recombination rate for a set of 1 , 000 coalescent model simulations . The expected b ¯ 1 of the barcode ensemble is informative of the recombination rate , growing monotonically with the later . Compared to sequentially Markov coalescent ( SMC ) approaches for ARG inference [34] , b ¯ 1 is strongly correlated with the number of recombinations in SMC ARGs derived from the same set of sequences ( Pearson’s r = 0 . 93 , p < 10−100 , S1 Fig ) . Although the coefficient of variation is ∼ 35% larger for b ¯ 1 ( S1 Fig ) , its computing time is substantially lower ( > 9 times faster after parallelizing in a modern 8-cores desktop computer , S1 Fig ) , being a robust approach to coalescent-model recombination rate estimation in large datasets . Furthermore , unlike the number of recombinations in SMC ARGs , b ¯ 1 is unbiassed at small recombination rates , vanishing when the recombination rate is zero ( Fig 11A ) . Although recombination rate estimation is a very direct example , the barcode ensemble of a sample of genetic sequences contains other rich phylogenetic information apart from b ¯ 1 , which can be used for more complex parameter estimation in structured models of evolution . Consider , for instance , the case of two divergent populations with migration and recombination discussed above . In this model , the average genetic distance between recombining sequences is expected to decrease with the migration rate , as the average time to the most recent common ancestor between foreign and local gametes in a population is shorter . In Fig 11B we show the dependence of the average death time ( 〈ϵd〉 ) on the migration rate parameter , for the barcode ensembles of a set of 900 coalescent model simulations with fixed recombination and variable migration rates . As expected , 〈ϵd〉 is informative of the migration rate , decreasing monotonically with the later . It is therefore a good measure for estimating migration rates . Consistently , 〈ϵd〉 correlates with time to the most recent common ancestor of recombining sequences in SMC ARGs obtained from the same data ( Pearson’s r = 0 . 55 , p < 10−72 , S1 Fig ) . Although the coefficient of variation of 〈ϵd〉 is ∼ 60% larger ( S1 Fig ) , extracting this type of information from SMC ARGs requires the implementation of a greedy algorithm , substantially increasing the running time ( ∼ 8 times slower in a single core of modern desktop computer , S1 Fig ) and therefore limiting its applicability to large datasets . These two simple examples illustrate the utility of barcode ensembles for building parameter estimators in specific models of evolution . Importantly , being model-independent , they are robust and flexible tools which can be applied in an infinitely large number of possible evolutionary models .
As the famous title of the essay by Dobzhansky “Nothing in Biology Makes Sense Except in the Light of Evolution” underscores , evolutionary processes are central orchestrating themes in biology . Mutations , recombinations and other evolutionary processes get imprinted into genomes through selection , reflecting the accumulated history giving rise to an organism . Phylogenetics try to reconstruct the evolutionary history through the comparison of genomes of related organisms . In addition to reporting relationships and elucidating particular histories , one would like to understand and quantify how different evolutionary processes have occurred . The identification and quantification of evolutionary processes can be challenging due to the lack of a well-established universal framework to capture evolutionary relationships beyond trees . In addition , robust statistical inference needs to exploit the large number of genomes that are now becoming available , aggravating the computational burden and obscuring interpretations . Ideally , we would like to have a biologically interpretable framework able to quantify different evolutionary processes by analyzing large numbers of genomes . In this paper we have proposed a few steps in this direction . We have extended the notion of barcodes in persistent homology to identify the genetic scale and number of recombination events . We have shown that , by correctly studying persistent homology in subsets of segregating sites , it is possible to characterize the genomic regions where recombination takes place and identify the gametes involved in particular recombination events . The persistent homology barcodes derived from each of these sets can be structured as a “barcode ensemble” where each bar captures a recombination event . Barcode ensembles can be interpreted as counting and quantifying the scale of recombination events in a variation of Ancestral Recombination Graphs ( ARGs ) . Topological ARGs represent a summary of potential recombination histories that can explain the data . The method proposed , TARGet , is scalable to hundreds of genomes . As an alternative to some phylogenetic networks , barcode ensembles provide robust quantification of events , the distribution of genetic scales , computational scalability and interpretative graphs . Barcode ensembles are versatile in that they do not assume any specific model of evolution , providing explicit , interpretable summaries of the minimal set of recombination events required to explain a sample of genetic sequences . Here we have illustrated their use in several practical cases . However , the range of possible applications is unlimited . In some cases , it may be convenient to perform minor modifications to the approach described here . For instance , although in our exposition we have only made use of Hamming distance and binary sequences , the main concepts we have presented extend straightforwardly to other genetic distances . The use of these metrics can be particularly useful in cases with rapidly diverging samples or substantial mutational biases . In other cases , information about the ancestral and derived alleles for each character in the sample may be available . Although tARGs have no natural directionality , the inclusion of the ancestral sequence in the original sample may lead in those cases to more stringent bounds on R ¯ min , similarly to what occurs with other approaches to recombination inference [22] . Finally , more efficient integer linear programming algorithms , like the one of [33] , could in principle be also generalized to the computation of barcode ensembles .
We extended the construction of ref . [45] to persistent homology barcodes . From a geometric perspective , this corresponds to projecting the original space on sets of mutually orthogonal hyperplanes in the ambient hypercube , and computing persistent homology in each of those projections . For that aim , we need to establish an ordering relation on barcodes . Being sets of intervals , it is natural to take the maximum of two barcodes to be given by the one with largest L0-norm , namely largest b1 . If both barcodes have the same L0-norm , we may successively compare other norms ( e . g . other Lp-norms ) , until the tie is broken or , otherwise , one of the two barcodes is arbitrarily chosen . The algorithm of [45] is then generalized to persistent homology barcodes as follows: The barcode ensemble of S is the union barcode R 1 m that results from this algorithm . We implemented the algorithm in a publicly available multi-threaded software , TARGet , which is distributed under the GNU General Public License ( GPL v3 ) . The application is fully written in Python 2 . 7 , and relies on Dionysus C++ library for persistent homology computations ( http://www . mrzv . org/software/dionysus ) . Since considering all possible sequence partitions is unnecessary and computationally infeasible in most cases , we follow the strategy of ref . [45] and allow the user to limit the number of partitions by the maximum number of segregating characters within each subset of S ( specified by the command line option -s ) , and by the maximum distance between segregating characters in the subset ( specified by the command line option -w ) . In addition , we also allow the user to exclude from S segregating characters that are compatible ( namely , that satisfy the Hudson-Kaplan four-gamete test [44] ) with all the other characters in S ( specified by the command line option -e ) . For each genomic interval , a filtration of Vietoris-Rips complexes is constructed using Hamming distance and the persistent first-homology group is computed over Z2 . We performed 4 , 000 simulations of a sample of 40 sequences with 12 segregating sites , using the software ARGweaver [34] . The population was simulated using a coalescent infinite sites model with recombination . The population-scaled recombination rate , ρ , was randomly generated in each simulation , taking values from a uniform distribution between 0 and 110 . For each simulated sample , Myers and Griffiths lower bound RMG ≤ Rmin was computed using the software RecMin [45] , with parameters -s 12 -w 12 . Lower bounds b ¯ 1 ≤ R ¯ min were computed using our application TARGet , with parameters -s 12 -w 12 . To study the dependence of b ¯ 1 on the recombination rate parameter in coalescent models , we performed 1 , 000 simulations of a sample of 200 sequences . The population-scaled recombination rate , ρ , was randomly generated in each simulation , taking values between 0 and 216 . For each simulated sample , TARGet was run with parameters -s 11 -w 11 , and ARGweaver’s tool arg-sample was run with parameters -m 7e-9 -n 400 --sample-step 10 , discarding the first 200 iterations . Samples of genetic exchange between two divergent populations were simulated using the software ms [59] , using the commands ms 300 1 -s 300 -r 40 10000 -I 2 250 50 -ej 6 . 0 1 2 -n 2 0 . 2 -m 1 2 0 . 5 and , ms 300 1 -s 300 -r 40 10000 -I 2 250 50 -ej 6 . 0 1 2 -n 2 0 . 2 respectively for the cases with and without migration . The barcode ensemble of each sample was computed using TARGet with parameters -s 12 -w 14 -e . To study the dependence of 〈ϵd〉 on the migration rate in this scenario , we performed 900 simulations using the software ms [59] and seq-gen [60] , with the commands ms 150 1 -T -r 60 10000 -I 2 125 25 -ej 6 . 0 1 2 -n 2 0 . 2 -m 1 2 X and seq-gen -mHKY -l 10000 -s 0 . 004 -p 50000 where the migration rate X in the first command takes random values from a uniform distribution between 0 and 2 . For each simulated sample , TARGet was run with parameters -w 8 -s 8 , and arg-sample was run with parameters -m 1e-7 -n 400 -r 1 . 5e-7 . We extracted from SMC ARGs the time to the most recent common ancestor of recombining sequences using a greedy algorithm that searches for the shortest non-zero path connecting the two sequences . We downloaded phased genotype data from HapMap phase III [53] , corresponding to all SNPs of LWK population between rs6457661 and rs3129301 in chromosome 6 . We also downloaded phased genotype data from 1 , 000 Genomes Project [3] , corresponding to all SNPs of LWK population between positions 32 , 887 , 978 and 32 , 927 , 978 of chromosome 6 , and half of the SNPs of LWK population between positions 234 , 190 , 031 and 234 , 291 , 193 of chromosome 1 . All coordinates refer to human assembly hg18 . The barcode ensemble of each dataset was computed using TARGet with parameters -s 12 -w 12 . Raw paired-end reads from 112 Darwin finches [51] were obtained from SRA archive ( accession number PRJNA263122 ) and aligned against the consensus sequence of Geospiza Fortis , version GeoFor_1 . 0/geoFor1 , scaffold JH739904 . We followed essentially the same procedure than that of ref . [51] for the alignment , SNP calling , genotyping and filtering . In short , the alignment was performed with Burrows-Wheeler aligner ( BWA ) [61] , version 0 . 7 . 5 , using BWA-MEM algorithm and default parameters . PCR duplicates were marked using Picard tools ( http://picard . sourceforge . net/ ) . Indel realignment , SNP discovery and simultaneous genotyping across the 112 samples was performed using Genome Analysis Toolkit ( GATK ) [62] , following GATK best practice recommendations [63] . SNP calls were filtered by keeping variants with SNP quality > 100 , total depth of coverage > 117 and < 1750 , ratio between SNP quality and depth of coverage > 2 , Fisher strand bias < 60 , mapping quality > 50 , mapping quality rank > -4 and read position rank sum > -2 . In total , 13 , 980 variant positions passed these filters . To avoid phasing errors , we only considered SNPs that were homozygous across the 120 samples . The resulting genotypes were processed with TARGet for barcode ensemble computation , using the options -s 14 -w 14 .
|
Evolution occurs through different mechanisms , including point mutations , gene duplication , horizontal gene transfer , and recombinations . Some of these mechanisms cannot be captured by tree graphs . We present a framework , based on the mathematical tools of computational topology , that can explicitly accommodate both recombination and mutation events across the evolutionary history of a sample of genomic sequences . This approach generates a new type of summary graph and algebraic structures that provide quantitative information on the evolutionary scale and frequency of recombination events . The accompanying software , TARGet , is applied to several examples , including migration between sexually-reproducing populations , human recombination , and recombination in Darwin’s finches .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"infographics",
"taxonomy",
"genome",
"evolution",
"phylogenetics",
"data",
"management",
"mathematics",
"algebra",
"genome",
"analysis",
"dna",
"homologous",
"recombination",
"computer",
"and",
"information",
"sciences",
"genomics",
"molecular",
"evolution",
"evolutionary",
"systematics",
"evolutionary",
"genetics",
"biochemistry",
"data",
"visualization",
"nucleic",
"acids",
"graphs",
"genetics",
"algebraic",
"topology",
"biology",
"and",
"life",
"sciences",
"topology",
"physical",
"sciences",
"dna",
"recombination",
"evolutionary",
"biology",
"computational",
"biology"
] |
2016
|
Inference of Ancestral Recombination Graphs through Topological Data Analysis
|
Solomon Islands is intensifying national efforts to achieve malaria elimination . A long history of indoor spraying with residual insecticides , combined recently with distribution of long lasting insecticidal nets and artemether-lumefantrine therapy , has been implemented in Solomon Islands . The impact of these interventions on local endemicity of Plasmodium spp . is unknown . In 2012 , a cross-sectional survey of 3501 residents of all ages was conducted in Ngella , Central Islands Province , Solomon Islands . Prevalence of Plasmodium falciparum , P . vivax , P . ovale and P . malariae was assessed by quantitative PCR ( qPCR ) and light microscopy ( LM ) . Presence of gametocytes was determined by reverse transcription quantitative PCR ( RT-qPCR ) . By qPCR , 468 Plasmodium spp . infections were detected ( prevalence = 13 . 4%; 463 P . vivax , five mixed P . falciparum/P . vivax , no P . ovale or P . malariae ) versus 130 by LM ( prevalence = 3 . 7%; 126 P . vivax , three P . falciparum and one P . falciparum/P . vivax ) . The prevalence of P . vivax infection varied significantly among villages ( range 3 . 0–38 . 5% , p<0 . 001 ) and across age groups ( 5 . 3–25 . 9% , p<0 . 001 ) . Of 468 P . vivax infections , 72 . 9% were sub-microscopic , 84 . 5% afebrile and 60 . 0% were both sub-microscopic and afebrile . Local residency , low education level of the household head and living in a household with at least one other P . vivax infected individual increased the risk of P . vivax infection . Overall , 23 . 5% of P . vivax infections had concurrent gametocytaemia . Of all P . vivax positive samples , 29 . 2% were polyclonal by MS16 and msp1F3 genotyping . All five P . falciparum infections were detected in residents of the same village , carried the same msp2 allele and four were positive for P . falciparum gametocytes . P . vivax infection remains endemic in Ngella , with the majority of cases afebrile and below the detection limit of LM . P . falciparum has nearly disappeared , but the risk of re-introductions and outbreaks due to travel to nearby islands with higher malaria endemicity remains .
Nations in the Southwest Pacific have endured considerable malaria transmission , with the highest Plasmodium falciparum burden outside the African continent and possibly the highest Plasmodium vivax transmission in the world [1] . Historically , transmission has ranged from hyperendemic areas in West Papua ( Indonesia ) and Papua New Guinea [2] to high and moderate transmission in Solomon Islands and Vanuatu [3] , which are the southwestern boundary of global malaria transmission . Intensified control over the last 20 years has resulted in remarkable declines in malaria transmission in this region [3 , 4] , reviving the agenda of elimination . However , it is in these countries where outstanding progress towards elimination has been made , that more knowledge is needed if the vision of malaria elimination is to be realized , such as reliable prevalence estimates , role of low-density , asymptomatic carriers and determinants of transmission maintenance . In Solomon Islands , the incidence of clinical malaria cases diagnosed by light microscopy ( LM ) dropped by 90% from 442/1000 population in 1992 [5] to 44/1000 population in 2012 [6] . These drops in incidence are similar to those achieved by the Malaria Eradication Program in Solomon Islands ( 1970–1975 ) [7] . National statistics based on passive surveillance indicate that 65% of clinical malaria cases in 2012 were attributable to P . falciparum , 33% to P . vivax and 2% to mixed P . falciparum/P . vivax . Conversely , active case detection surveys indicate that P . vivax is the predominant species in the general population [6] . Current malaria transmission appears to be focal , ranging from moderate to high levels in Honiara City ( 96/1000 ) and Guadalcanal ( 64/1000 ) to very low in Temotu ( 10 . 8/1000 ) and Isabel provinces ( 1 . 2/1000 ) . Temotu and Isabel are the only two provinces in which pilot elimination agenda has been proposed to be actively pursued , having resulted in more intensive control activities and interventions including stratification , active case detection , and the earlier roll out of control activities ( e . g . rapid diagnostic tests , RDTs and indoor residual spraying ) than the rest of the country [3] . These provinces are also the only areas of Solomon Islands with recent surveys in which both LM and PCR-based diagnoses of Plasmodium spp . infections were performed [8 , 9] . In 2008 , a parasite prevalence of 2 . 7% by LM was found in Temotu , with P . vivax accounting for 82 . 5% of infections . Only 5 . 5% of these infections were associated with febrile illness . Among a subset of 1 , 748 samples , which included LM positive , febrile and 10% of LM negative participants , an additional 63 P . falciparum , 23 P . vivax and 10 mixed P . falciparum/P . vivax infections were detected by PCR , indicating a 6 . 5% prevalence of sub-microscopic infections . Even lower levels of infection were reported in Isabel in 2009: 1 of 8 , 554 participants had a LM-detectable P . falciparum infection ( 0 . 01% ) . In a random subset of 2001 participants , PCR identified an additional 13 ( 0 . 55% ) P . vivax infections . PCR consistently detects at least twice as many infections as LM [10] . Numerous studies have confirmed that sub-microscopic infections are a common feature of malaria endemic areas , spanning all age groups and involving both P . falciparum and P . vivax [11–13] . Although these sub-microscopic infections are rarely associated with febrile illness , they have been shown to be efficient gametocyte producers [14–19] and thus constitute a source of ongoing transmission [10] . Given the lack of data from other areas of Solomon Islands , it is currently unknown whether the pattern of asymptomatic , low-density infection carriage identified in Temotu and Isabel [8 , 9] is unique to these elimination provinces . In addition , whereas these earlier surveys detected a large burden of sub-microscopic infections , they did not determine if these infections were also gametocytaemic and therefore did not assess their potential contribution to transmission . Therefore , we conducted in May-June 2012 a household-based , cross-sectional survey in Ngella , Central Islands Province to determine how common low-density , asymptomatic infections are in communities where transmission is mesoendemic and whether these infections are gametocyte producers and hence , potential contributors to local transmission . This survey is the first epidemiological description of malaria in Ngella since the 1970–1975 Malaria Eradication Program [7] and the only one in Solomon Islands to employ highly sensitive molecular diagnosis for the detection of both blood-stage parasites and gametocytes .
This study was approved by The Walter and Eliza Hall Institute Human Research Ethics Committee ( HREC number 12/01 ) and the Solomon Islands National Health Research Ethics Committee ( HRC12/022 ) . The informed consent process recognized the community and cultural values of Solomon Islands . Following consultation with and approval by community leaders , community meetings were held to explain the aims , risks and potential benefits of the study . Individual informed consent was obtained from all participants or the parent or legal guardian of children<18 years of age . At the point of collection , all samples were de-identified . Ngella , previously known as the Florida Islands , consists of 3 islands , Anchor , Big Ngella and Small Ngella , located approximately 27 miles north of Guadalcanal and 50 miles southwest of Malaita ( Fig 1 ) . Along with Tulaghi , Savo , Russel and Buenavista Islands it forms part of the Central Islands Province ( Fig 1 ) . Despite their proximity , the three islands of Ngella have diverse geographical characteristics: Anchor Island is characterized by less dense rainforest and sandier soil . Big Ngella is heavily forested , although commercial deforestation is common , and smaller villages are encountered in the Bay area around Tulagi , the provincial capital . The more remote northern villages of Big and Small Ngella and those on the southern coast are larger . The communities of the Utuha Channel lay in an extensive mangrove system and are smaller in size . There is minimal seasonal variation in temperature and despite a northwesterly monsoon from November-April , the distinction between wet and dry season is not pronounced . The most recent census estimates 26 , 051 inhabitants ( approximately 60% of these reside in Ngella ) , 49% females and a median age of 19 . 9 years [20] . There is significant migration between Ngella and other malaria endemic areas , in particular Honiara ( Guadalcanal ) and Malaita provinces . These provinces are well connected to Ngella by a popular ferry service and numerous private , unscheduled motorized boat trips . The Ngella population is serviced by a hospital in Tulagi , six rural health sub-centres and ten nurse aid posts . National malaria statistics describe Ngella as mesoendemic , with a reported Annual Parasite Index [21] of 46 . 1/1000 in 2012 , P . falciparum being the main cause of malaria cases [6] . Overall API for Solomon Islands indicates that there were two transmission peaks in 2012 for the months of February and October . As elsewhere in the country , long lasting insecticidal nets and indoor residual spraying are the mainstay of malaria control in Ngella . Cases are diagnosed by LM or RDT and treatment with artemether-lumefantrine has been introduced nationally in 2008 . The last malaria epidemiological report of Ngella [7] described it as ‘the most malarious group in all Solomon Islands” and the “most difficult from which to clear malaria” . Malariometric surveys preceding the Malaria Eradication Program ( March 1965—January 1970 , unpublished World Health Organisation Field Reports ( reviewed in [8] ) identified a combined parasite rate of 69 . 6% and a spleen rate of 69 . 3% in the 2–9 years age group . In the same surveys , villages on the North coast had spleen rates in the 80% range and qualified for the hyperendemic classification [7] , whereas the villages in the Bay area and South coast were noted to have had spleen rates in the 30–50% range [7] . A representative population sample was obtained with a household-based sampling strategy of villages in 5 distinct geographical regions ( Fig 1B , Anchor Island , North Coast , Bay , South Coast and Channel ) . The survey included 3501 individuals of all ages ≥6 months residing in 874 households in 19 randomly selected communities . The households were enumerated and geo-positioned and demographic information of the male and female heads of the household collected . Enumeration , but not geopositioning , was achieved for the households in the villages of the South Coast . The timing of the survey was approximately 4 weeks after the peak of the wet season . Following consent and enrolment from each participant , a short clinical assessment was conducted ( including tympanic temperature , history of fever during the previous 48 hours , history of malaria in the last 2 weeks and spleen size in children 2–9 years old ) and demographic information collected ( age , sex , residency status and history of travel , bed net use ) . A febrile participant was defined as having a tympanic temperature ≥38 . 0°C and/or a history of febrile illness in the past 48 hours . Study participants who reported being ill at the time of the survey were diagnosed by RDT ( Access Bio , CareStart , USA ) and treated if positive with artemether/lumefantrine , as per the national treatment guidelines . Where available the participant’s health records were checked for recent anti-malarial treatment and applicable information recorded . A 250 μL finger prick blood sample was collected into EDTA-Microtainer tubes ( Becton Dickinson , NJ , USA ) . 50 μL were immediately stabilized in 250 μL RNAProtect ( Qiagen , Germany ) for RNA studies and stored at ice pack cooling conditions until their transport to a centralized field laboratory . Thick and thin films were prepared for determination of microscopic malaria infection . Haemoglobin measurement was performed with Hemocue HB 301 analyzer . A measurement below 11g/dL was classified as anaemia . Upon return to the centralized laboratory , the RNAProtect fractions were frozen immediately . The remaining 200 μL of whole blood was separated into red blood cells pellets and plasma and promptly frozen . Giemsa stained blood films were examined under x1000 power . One hundred fields of view were examined before calling a sample “no parasites seen” . When a parasite was observed , counts of both white cells and parasites were commenced , and continued until 300 white cells had been counted . The parasite count was then calculated , based on an assumed white cell count of 8 , 000 white cells/ μL . However if no further parasites were observed , the process of scanning to a total of 100 fields of view was completed . When only 1 parasite had been observed in 100 fields of view , an assumed count at the notional lower limit of detection of 10 parasites/μL was applied , based on a further assumption of an average of 8 white cells per field of view . All slides were stained within 24 hours at the regional malaria laboratory and read by experienced microscopists , all of whom had completed WHO quality assurance courses . All LM positive slides as well as the slides from all PCR positive / LM negative plus 10% of LM & PCR negative slides were re-read by an Australian Level 1 expert microscopist that was blinded to the PCR results . None of the 10% LM negative slides were found to be positive by the expert microscopist . In case of discrepancies between the two microscopy reads , the read of the expert microscopist was considered final . Genomic DNA ( gDNA ) was isolated from red blood cell pellets ( 100 μL , corresponding to 200 μL whole blood ) using FavorPrep 96-well Genomic DNA kit ( Favorgen , Taiwan ) . DNA was eluted in 200 μL elution buffer and stored at -20°C . The RNA isolation procedure from whole blood in RNAProtect cell reagent has been described elsewhere [22] , the only exception being an increased elution volume of 60μL of RNase-free water . Due to problems with storage of RNAProtect samples in the field , the quality of the RNA was tested using an RT-qPCR for the human beta globin transcript [23] . This revealed a 10x lower total human RNA concentration than in samples from a comparable study in Papua New Guinea [24] . RNA samples were therefore concentrated 10-fold using a CentriVap Concentrator ( Labconco , United States ) before testing for the presence of gametocytes . All 3501 DNA samples were first screened using a genus-specific qPCR targeting a conserved region of the 18S rRNA gene [22] . Singleplex species-specific P . falciparum and P . vivax Taqman qPCRs and a duplex P . malariae/P . ovale qPCR , targeting species-specific regions of 18S rRNA gene , were used to identify species as described previously [22 , 25] . Prevalence values reported in this study include only those infections confirmed by the species-specific qPCR Taqman assays . Each detection experiment carried a dilution series of plasmids containing the target sequence of each PCR ( 104 , 103 , 102 , 101 , 5 , 100 copies/μL ) , in duplicate , and were used to determine standard curves and therefore estimate parasite densities ( reported as 18S rRNA gene copy numbers/μL ) . All assays were run in 384-well plate format on the Roche LightCycler480 platform . Those infections detected by qPCR , but not by LM , were defined as sub-microscopic infections . P . falciparum and P . vivax samples that were positive by species-specific Taqman qPCR were examined for presence of gametocytes using RT-qPCRs targeting the pfs25 and pvs25 orthologues , which are expressed only in mature gametocytes , as described previously [22] . All gametocyte assays were also run in 384-well plate format on the Roche LightCycler480 platform . All samples that were P . falciparum or P . vivax positive were genotyped to determine the multiplicity of infection ( MOI ) using highly diverse size-polymorphic molecular markers msp2 for P . falciparum and msp1F3 and MS16 for P . vivax , respectively . PCR and capillary electrophoresis were performed with slight modifications to the published protocols [26 , 27] . Genotyping data was analyzed as described previously [26 , 27] . Study data were collected and managed using REDCap electronic data capture tools hosted at the Walter and Eliza Hall Institute [28] . Analyses were done using the STATA12 statistical software package ( College Station , TX ) . Differences in participant characteristics at enrolment and prevalence differences among geographical areas and groups of individuals were assessed using Chi-square ( χ2 ) or Fisher’s exact tests . Differences in median ages and median household size were explored with quantile regression . Univariable and multivariable logistic regression were used for associations of P . vivax infection and exposure variables . Associations with P . vivax parasite density were investigated in simple and multivariable linear regression models on only those subjects who tested positive to qPCR diagnosis . Poisson regression analyses were utilized to explore associations between multiplicity of infection and exposure variables .
A total of 3501 Ngella residents across 874 households were surveyed . The gender and age profiles of the participants were representative of the Central Islands Province population , with 52 . 5% females and a predominance of younger individuals ( median age 18 years ) . The age distribution was as follows: <2 years , 4 . 7%; 2–4 years , 10 . 6%; 5–9 years , 14 . 8%; 10–14 years , 14 . 3%; 15–19 years , 7 . 5%; 20–39 years , 27 . 0%;>40 years , 21 . 3% . The majority of participants ( 95 . 2% ) resided in the village for ≥ 2 months . Of 447 participants who spent at least one night outside their village of residence in the last month , 69 . 1% travelled within Central Islands Province . 73 . 3% of participants reported having slept under a long lasting insecticidal net the night before and 56 . 4% owned a bednet for longer than 24 months . Of all households , 84 . 5% of households reported to have been sprayed with insecticide , and 70 . 4% of household heads spoke English . Of all participants , 687 ( 19 . 4% ) had a history of fever in the previous two days , 685 ( 19 . 7% ) reported feeling unwell/sick at the time of survey and 23 . 3% had a haemoglobin measurement <11g/dL . No participant aged 2–9 years of age was found to have an enlarged spleen . A detailed description of demographic and clinical characteristics by geographical region is given in S1 Table . Overall , 130 individuals ( 3 . 7% ) had Plasmodium spp . parasites detectable by LM: 126 P . vivax , three P . falciparum mono-infections and one P . vivax/P . falciparum mixed infection . No infections with P . malariae or P . ovale were observed . The prevalence of P . vivax infection varied significantly by geographical region ( p<0 . 001 ) ( Fig 1B ) and was lowest in the South Coast and Anchor regions ( 0 . 8% ) , followed by Channel ( 3 . 0% ) and Bay ( 3 . 5% ) and North Coast ( 11 . 7% ) . P . vivax prevalence showed strong age trends and peaked in adolescents 10–15 years of age ( 8 . 6% , p<0 . 001 ) ( Fig 2A ) . Overall , 468 participants ( 13 . 4% ) had qPCR-detectable infections: 463 were P . vivax mono-infections and five were mixed P . falciparum/P . vivax infections ( 0 . 14% ) . The 126 P . vivax infections and the one mixed infection by LM were confirmed by qPCR . Overall , 72 . 9% of P . vivax infections were sub-microscopic . In two of the catchments , Kelarekeha and Vuturua ( Fig 1B ) , only sub-microscopic infections were observed among 59 and 156 individuals surveyed , respectively . P . vivax qPCR prevalence displayed spatial heterogeneity among the five geographical areas and the 19 catchments , varying from 3 . 0–38 . 5% . Prevalence by qPCR was highest in villages on the North Coast ( 25 . 0–38 . 5% ) and lowest on Anchor Island ( 3 . 0–5 . 5% ) ( Fig 1B ) . Of 874 households sampled across Ngella , 559 had no infected members , 210 had only one infected member and 105 had two or more infected members . There was no association between household size and probability of being infected ( p = 0 . 550 ) . Not taking into account any other variables , there was an increased risk of being infected if at least one other member of the household was infected ( OR = 2 . 59 , p<0 . 001 , CI95[2 . 13 , 3 . 16] ) . P . vivax prevalence was age-dependent ( p<0 . 001 ) , lowest in<2 years ( 3 . 0% , n = 166 ) and peaking in the 10–14 year old age group ( 24 . 3% , n = 499 ) ( Fig 2A ) . Prevalence of infection did not differ significantly between male and female participants ( p>0 . 650 ) . Participants who were residents ( lived in the village ≥ 2 months ) were more frequently infected with P . vivax than non-residents ( infected residents: 13 . 7% vs . infected non-residents 8 . 3% , p = 0 . 045 ) . Once residency status was taken into account , recent travel ( defined as spending at least 1 night away from the village of residence in the last month ) was not associated with a difference in infection risk ( p = 0 . 300 ) . Those living in a household where the household head speaks English , a proxy for education level , were infected less frequently ( 12 . 0% ) than those living in a household where the head does not speak English ( 18 . 5% , p<0 . 001 ) . There was a moderate increase in risk of P . vivax infection in those who reported not having slept under a net the night before compared to net users ( users: 12 . 6% vs . non-users: 15 . 6% , p = 0 . 022 ) . The majority of P . vivax-infected individuals ( 84 . 6% ) neither reported febrile symptoms ( defined as history of fever or measured fever at survey ) nor feeling ill ( 85 . 4% ) . Six of the 26 participants that had a measured fever at the time of the survey ( tympanic temperature ≥38°C , 18 . 8% ) were infected with P . vivax . Compared to uninfected participants , those with a P . vivax infection were less likely to report having had febrile symptoms in the previous two days ( uninfected 20 . 0% vs . infected 15 . 2% , p = 0 . 014 ) or report feeling unwell at the time of survey ( uninfected 20 . 5% vs . infected 14 . 6% , p = 0 . 003 ) . A total of 280 P . vivax infections ( 60 . 0% ) were both asymptomatic and sub-microscopic . There were no significant differences in the proportion of asymptomatic P . vivax infections between different age groups ( p> 0 . 200 ) and regions ( p> 0 . 240 ) . Of 468 P . vivax-infected individuals , 19 . 6% had a haemoglobin<11 g/dL compared to 23 . 9% of uninfected individuals ( p = 0 . 045 ) . Age was the strongest independent association with P . vivax . infection , peaking in 10–14 year olds ( Table 1 , AOR = 8 . 41 , p<0 . 001 ) and remaining relatively steady for subjects aged 15 years and above ( Table 1 , 15–19 years AOR = 5 . 03; 20–39 years AOR = 4 . 75; ≥ 40 years AOR = 4 . 40; p<0 . 001 ) . The reference group in this analysis was composed of children aged <2 years . Being a local resident , region of residency and living in a household with at least 1 additional infected member were all associated with an excess risk of infection . Significant protective factors included an English-speaking household head and reporting feeling unwell at the time of the survey . Detailed results are given in Table 1 . Parasite densities by LM count were low ( estimated geometric mean 24 . 3 parasites/μL , CI95 [19 . 2 , 30 . 8] ) ( Fig 2B ) . Of 130 LM-detectable infections , 55% ( n = 70 ) , were at the assumed limit of practical detection , i . e . approximately 10 parasites/μL . Similarly , P . vivax parasite densities by qPCR were also low ( estimated geometric mean 18S DNA copy numbers of 4 . 6/μL , CI95 [3 . 9 , 5 . 4] ) ( Fig 2B ) . In LM and qPCR positive infections , parasite density by LM ( parasites/μL ) and by qPCR ( 18S DNA copies/μL ) were correlated ( n = 130 , R2 = 0 . 76 , p<0 . 001 ) . Factors predicting parasite densities by qPCR included age , history of fever in the preceding two days , anaemia ( haemoglobin<11 g/dL ) and geographical region of sampling . Parasite densities were highest among individuals aged <2 years and decreased in older age groups ( p<0 . 001 ) . Detailed results of associations with P . vivax density are given in Table 2 . Genotyping results for markers msp1F3 and/or MS16 were obtained from 349 P . vivax positive samples . Both markers were highly diverse; 15 msp1F3 alleles and 43 MS16 alleles were detected ( Fig 3 ) , resulting in an expected heterozygosity ( HE ) of 0 . 834 for msp1F3 and 0 . 937 for MS16 . Out of 349 samples , 102 ( 29 . 2% ) carried multi-clonal infections . MOI ( combined msp1F3 and MS16 ) , defined as the concurrent infections per individual , ranged from 1 to 4; mean MOI was 1 . 36 . Mean MOI did not differ significantly among age groups ( Fig 2D , p = 0 . 774 ) or among geographical regions ( p = 0 . 610 ) . P . vivax-infected individuals living in a household with at least one other infected individual had moderately higher mean MOI ( 1 . 56 ) than those who were the sole infected person of the household , but evidence for an association was moderate to weak ( mean MOI = 1 . 42 , p = 0 . 086 ) . Mean MOI was positively associated with qPCR parasite density ( p = 0 . 007 ) . In 110 of 468 ( 23 . 5% ) P . vivax infections , gametocytes were detected by the pvs25 RT-qPCR , resulting in a population gametocyte prevalence of 3 . 14% . More gametocytes were detected in LM-positive ( 41 . 5% ) than sub-microscopic infections ( 16 . 6% , p<0 . 001 ) . The proportion gametocyte positive ( defined as percentage of gametocytaemic P . vivax carriers ) was highest in the 2–9 year olds ( 39 . 8% , n = 98 ) and decreasing sharply in the 10–14 year olds ( 20 . 7% , n = 121 , p>0 . 001 ) . The lowest proportion gametocyte positive was found among the P . vivax carriers in the 15–19 years and>40 years age groups ( 15 . 8% , n = 38 and 15 . 9% , n = 88 , respectively ) ( Fig 2C ) . By qPCR , P . falciparum was detected in 5 individuals , all of whom were co-infected with P . vivax . Of these , only four infections were detectable by LM , one as a co-infection with P . vivax and three as P . falciparum mono-infections . In two of the LM mono-infections and the mixed infection , only P . falciparum gametocytes were observed on the blood smear . The range of parasite densities , by qPCR , was 7 . 35–364 copy numbers/μL and in LM positive samples the densities ranged from 20 to 1430 parasites/μL . The low number of P . falciparum infections precluded analyses with densities for this species . The presence of gametocytes in the four LM-positive infections was confirmed by pfs25 RT-qPCR . No gametocytes were detected in the one sub-microscopic P . falciparum infection by RT-qPCR . All five P . falciparum-infected individuals resided in the same village ( Halavo , circled in Fig 1B ) and ranged in age from 3 to 60 years . The oldest had a history of fever in the preceding two days . Two of the carriers had a haemoglobin measurement <11g/dL . None of the individuals reported to have slept outside the village in the previous month . All five P . falciparum infections were monoclonal and carried the same msp2 genotype of the Fc27 subtype .
Solomon Islands has achieved a remarkable 90% reduction in malaria incidence over the last two decades as a result of scaled-up malaria control interventions [6] and is now intensifying its efforts towards malaria elimination [3] . The present study is the first to undertake sensitive molecular diagnosis at this scale in Solomon Islands and the first large epidemiological description of malaria in Ngella since the Malaria Eradication Program ( 1970–1975 ) . Our findings illustrate a striking distinction between the epidemiology of P . falciparum and P . vivax in Ngella . High prevalence ( 13 . 4% by qPCR ) and genetic diversity , as well as an increased risk for local residents and evidence of potential within-household transmission indicate considerable levels of endemic P . vivax transmission . There was significant variation of P . vivax transmission in different regions of Ngella , with the highest prevalence found on the remote North Coast ( 25 . 0–38 . 5% ) , which prior to the Malaria Eradication spraying operations was described as holoendemic and having an environment highly favourable to the mosquito [7] . The lowest rates of P . vivax infection were observed on Anchor ( 3 . 9% by qPCR ) , where 15 years ago a community-based initiative eliminated a substantial number of breeding sites through environmental management [Lodo , personal communication] . It is therefore likely that the presence of suitable larval habitats and vector abundance may be key factors influencing P . vivax transmission on Ngella . It remains unclear whether autochthonous P . falciparum transmission remains in Ngella or parasites are being re-introduced by incoming travelers or returning residents from areas with higher P . falciparum burden , such Guadalcanal or Malaita provinces . In this survey , only five P . falciparum cases were identified in the village of Halavo ( Fig 1B ) . As all five infections carried the same msp2 allele and four were gametocytaemic , a small local outbreak following recent re-introduction seems more likely . This is reminiscent of the situation in epidemic-prone areas of the Papua New Guinea highlands , where a clonal P . falciparum epidemic on a background of endemic , low level P . vivax transmission has been reported [29] . P . falciparum populations in neighbouring Guadalcanal province were in fact found to be of low genetic diversity [30 , 31] . Based on case statistics at the local health facilities , 30% of malaria cases detected in Central Islands Province are caused by P . falciparum [6] indicating that either importation of P . falciparum parasites is common or that low levels of endemic P . falciparum transmission may remain in some parts of Ngella . Further studies are therefore required to ascertain the absence of endemic P . falciparum transmission in this area of Solomon Islands and whether the cases found are the result of inter-island travel . The current situation of malaria in Ngella ( i . e . 3 . 7% prevalence by LM , clear P . vivax dominance and absence of enlarged spleens in children 2–9yrs of age ) is a consequence of the dramatic reduction in malaria transmission achieved throughout Solomon Islands in the last 20 years [6] . This change is similar to that encountered at the end of 1974 , after approximately 5 years of twice-yearly Malaria Eradication Program spraying . Then , prevalence in 2–9 year olds had dropped from pre-spraying rates of 60% to 1 . 4% and P . vivax became predominant [7] . Similar shifts in malaria epidemiology were also observed in the elimination provinces of Temotu [9] and Isabel [8] . In Temotu , P . falciparum accounted for 17 . 5% of infections in population survey conducted in 2008 [9] , but by 2012 , the national program’s surveillance system reported only P . vivax cases from both Temotu and Isabel [6] . This shift in the relative importance of P . falciparum and P . vivax are not unique to Solomon Islands and have been reported after periods of sustained malaria control from other settings where P . falciparum and P . vivax occur sympatrically , such as the Amazon [21 , 32] , Central America [4] and Thailand [33] . As in other endemic settings [12 , 13 , 34] , P . vivax infections were of low density and PCR found three times more infections than LM . The majority of infections were not accompanied by febrile symptoms or anaemia . On the contrary , participants who reported feeling unwell or febrile were less likely to be infected with P . vivax . While this significantly lower level of febrile symptoms in P . vivax carriers is likely to be an artifact of the large samples size it does indicate that P . vivax is not a common cause of fever in Ngella . Whereas asymptomatic P . vivax infections have been commonly found in areas of high transmission [12 , 35 , 36] , the advent of molecular diagnosis has revealed that even at low transmission the majority of infections in cross-sectional surveys are symptomless [11 , 37 , 38] , including in the previous surveys in Temotu [9] and Isabel [8] where 97 . 1% and 92 . 9% of P . vivax infected individuals infections were asymptomatic , respectively . Both the presence of P . vivax infections and their level of parasitaemia were found to be strongly age-dependent , albeit in different ways: while P . vivax parasite densities decreased with age , prevalence of P . vivax infections rose throughout childhood and only started dropping in adolescents and adults . These contrasting patterns are most likely due to local mosquito biting behavior and acquisition of immunity . Anopheles farauti , the only coastal malaria vector in Solomon Islands , is biting predominantly in the early evening ( i . e . before 10pm ) and outdoors [39] , when small children tend to be indoors but older ones still active . The increase in prevalence during childhood is thus likely to represent an increase in exposure to infective bites . At all levels of transmission , immunity to P . vivax tends to be more rapidly acquired than that to P . falciparum [40] . Thus , the strong reduction in prevalence and parasite densities with increasing age in Ngella indicate that P . vivax transmission there remains sufficiently high for relatively rapid acquisition of clinical and anti-parasite immunity . Despite very low overall parasite densities , gametocytes were detected in almost a quarter of all P . vivax infections ( in 41 . 5% of LM-positive infections and 16 . 6% of sub-microscopic infections ) . Given issues with RNA quality , it is likely that the gametocytaemic reservoir in Ngella was underestimated in our survey and the true prevalence of gametocytes is higher , especially in the sub-microscopic group . Given the rapid and ongoing production of P . vivax gametocytes , most if not all , blood stage infections could harbor concurrent gametocytes [41] . Whilst sub-patent P . falciparum infections have been shown to infect up to 43 . 5% of mosquitoes [17 , 19] , the role of sub-microscopic P . vivax gametocyte carriage in sustaining transmission is poorly understood . The capacity of sub-microscopic P . vivax infections to infect mosquitoes has been established in studies from Thailand [18 , 42 , 43] , Sri Lanka [44] , Peru [45] and malaria therapy settings [14 , 15] , but at varying proportions and with weak associations of gametocyte density . Although sub-patent infections may infect fewer mosquitoes , their higher prevalence in endemic settings may mean that the net transmission potential of low-density infections is higher . In Ngella , asymptomatic , sub-microscopic infections of adolescents and adults may thus be an important source of local transmission . These considerations may constitute a significant challenge to the success of the Solomon Islands malaria control program . The national malaria surveillance system , based on passive case detection and irregular mass blood surveys , only employs traditional microscopy diagnosis . This diagnostic test may not only underestimate the true burden of malaria in the Solomon Islands but also lack the means to detect and attack a substantial part of the P . vivax transmission reservoir . Despite outstanding gains in the last two decades , traditional tools of the Solomon Islands malaria control program may therefore have reached their effectiveness in the face of a large and silent reservoir of P . vivax infection . Our observation that people living in a household with another P . vivax infected individual is a noteworthy finding . Not only does it indicate likely within-household transmission , but also highlights that reactive case detection strategies [46–48] and focal mass drug administration [34] might be appropriately applied in Solomon Islands . In the Southwest Pacific , MDA campaigns that included primaquine to target the undetectable liver stage parasites have previously been successful in interrupting P . vivax transmission on Aneytium Island in Vanuatu [49] and Nissan Island in Papua New Guinea [50] . Combining automated registration of observed cases and rapid identification of transmission foci ( e . g . in a spatial decision support system ) [51] with reactive mass-screen and treat ( MSAT ) or with focal , household-based mass drug administration [52 , 53] should therefore be evaluated as possible additional malaria elimination tools in Solomon Islands and neighbouring Vanuatu . All interventions will be most efficacious if they include routine administration of primaquine to all P . vivax infected individuals . This will however require addressing the challenges posed by the potential primaquine toxicity in G6PD deficient individuals .
|
Solomon Islands , an island nation in the Southwest Pacific that has seen dramatic reductions in malaria transmission over the past 20 years , is aiming for malaria elimination . There is an increasing recognition that a substantial reservoir of asymptomatic and often sub-microscopic Plasmodium spp . infections exists even in low transmission settings . However , the potential role for these infections in sustaining transmission and the difference in response of the two most common malaria parasites , P . vivax and P . falciparum , to intensified control remains unclear . In May-June 2012 , we therefore performed a cross-sectional survey of 3501 residents of all ages of Ngella , a low transmission area in Central Islands Province , to assess the prevalence of P . vivax and P . falciparum infection , determine the proportion of sub-microscopic and afebrile infections and evaluate whether gametocytaemic , and thus potentially infectious , individuals are present . Our survey showed a marked epidemiological contrast between P . vivax and P . falciparum . Although prevalence varied significantly among different regions of Ngella , P . vivax remains firmly endemic , with high rates of sub-microscopic , afebrile and genetically diverse infections . The presence of gametocytes among both sub-microscopic and microscopy positive , asymptomatic infections indicates that these infections contribute significantly to sustaining P . vivax transmission . P . falciparum , on the other hand , appears to be more amenable to control interventions . Only five P . falciparum infected individuals were detected , and all were residents of the same village . These infections carried the same msp2 clone . This difference highlights the larger challenge of eliminating P . vivax compared to P . falciparum in areas where they are co-endemic . In particular , the challenge posed by the presence of a large reservoir of silent P . vivax infections will need to be addressed if control of this parasite is to be accelerated and elimination achieved .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
High Rates of Asymptomatic, Sub-microscopic Plasmodium vivax Infection and Disappearing Plasmodium falciparum Malaria in an Area of Low Transmission in Solomon Islands
|
The development and maintenance of polarized epithelial tissue requires a tightly controlled orientation of mitotic cell division relative to the apical polarity axis . Hepatocytes display a unique polarized architecture . We demonstrate that mitotic hepatocytes asymmetrically segregate their apical plasma membrane domain to the nascent daughter cells . The non-polarized nascent daughter cell can form a de novo apical domain with its new neighbor . This asymmetric segregation of apical domains is facilitated by a geometrically distinct “apicolateral” subdomain of the lateral surface present in hepatocytes . The polarity protein partitioning-defective 1/microtubule-affinity regulating kinase 2 ( Par1b/MARK2 ) translates this positional landmark to cortical polarity by promoting the apicolateral accumulation of Leu-Gly-Asn repeat-enriched protein ( LGN ) and the capture of nuclear mitotic apparatus protein ( NuMA ) –positive astral microtubules to orientate the mitotic spindle . Proliferating hepatocytes thus display an asymmetric inheritance of their apical domains via a mechanism that involves Par1b and LGN , which we postulate serves the unique tissue architecture of the developing liver parenchyma .
The liver is a vital organ . Hepatocytes occupy more than 85% of the parenchymal liver mass and are responsible for a wide range of biological processes . These include the synthesis of plasma proteins and the processing of nutrients and toxic compounds from the blood that passes through the liver sinusoids . Hepatocytes also produce and secrete bile . Bile contributes to fat emulsion in the intestine and the elimination of detoxified compounds via the feces . Hepatocytes form a branching network of bile canaliculi between the cells that efficiently drains the secreted bile out of the liver parenchyme while keeping it separate from the blood [1] , [2] . The microanatomy of this canalicular network is unique to the liver [3] . Defects in the bile canalicular network and bile flow are associated with liver diseases [4] . Knowledge of the cell biological principles and molecular mechanisms that underlie the development of the bile canalicular network is limited . This is in part due to the lack of in vitro cell culture model systems that combine cell proliferation and canalicular network formation . Nevertheless , different in vitro cell model systems can reproduce specific steps in the process of bile canalicular network formation . For instance , from early microscopy studies of embryonic rat livers we know that the formation of isolated small spherical lumens between mitotically active hepatocytes is the first step in bile canalicular network development [5]–[8] ( Figure 1A ) , and this process is reproduced by hepatic HepG2 [9] , [10] and WIF-B9 [11] cell lines . Both in vivo and in vitro , the formation of these primordial intercellular lumens is accompanied by the segregation of the hepatocyte surface into a lumen-facing apical domain and a sinusoid-facing basal domain , each with a specific protein and lipid composition ( Figure 1A ) [7]–[11] . The establishment of cell surface domains is the hallmark of apical–basal cell polarity [12] . The early establishment of apical–basal polarity is instrumental for the functional shaping of a proliferating epithelial cell mass [13] , [14] . Indeed , dividing cells not only generate enough critical cell mass to create the tissue , but they also make use of their apical–basal polarity axis ( PA ) to orientate their mitotic spindle apparatus [15] . By orientating its mitotic spindle apparatus , the dividing polarized epithelial cell can control the position of the emerging new nuclei , and hence the position of the daughter cells , relative to the position of the primordial apical domain and lumen . The same principles are used when dividing cells repair tissue damage [16] . The unique microanatomy of the bile canalicular network suggests that the mode of cell division orientation in hepatocytes—from the moment that they have established apical–basal polarity—differs from that observed in “simple” epithelial cells such as intestinal or kidney epithelial cells . Indeed , simple epithelial cells do not develop a canalicular network between cells . Instead , they develop large cystic lumens and tubes ( Figure 1B ) via a process that is dependent on a Leu-Gly-Asn repeat-enriched protein ( LGN ) –mediated orientation of the mitotic spindle apparatus that is strictly perpendicular to the apical–basal axis , and the resultant symmetric segregation of the apical domain to both daughter cells [17]–[19] ( Figure 1D ) . A mitotic spindle orientation that is strictly perpendicular to the apical PA would in mitotic hepatocytes thus be predicted to promote the development of cystic lumens rather than of canalicular networks ( Figure 1C ) . In order to investigate the orientation of the mitotic spindle in hepatocytes within their native environment , immunohistochemistry can be performed on fixed liver tissues . To study the molecular regulation and the dynamics of mitotic spindle orientation and cell division in hepatocytes , cell lines are the best model of choice . In this study we have combined the analysis of liver tissues with that of HepG2 and WIF-B9 cell lines to investigate the relationship between cell polarity and the orientations of the mitotic spindle and cell division at the molecular level in hepatocytes .
Commercial antibodies used for immunofluorescence are listed in Table S1 . The rabbit anti-LGN antibody was described earlier [20] . Phalloidin-TRITC was used to label F-actin ( P1951; Sigma ) . DAPI was from Invitrogen , and DRAQ5 was purchased from Cell Signaling Technology . The plasmid expressing H2B-mCherry was a kind gift from B . Giepmans ( University Medical Center Groningen , the Netherlands ) . The plasmid expressing human Par1b was a kind gift from H . Miki ( Osaka University , Japan ) . Overexpression was obtained by cloning constructs into a lentiviral expression system [21] . Briefly , constructs were cloned into pENTR1A ( Addgene plasmid 17398 ) and recombined into pLenti-CMV-Puro ( Addgene plasmid 17452 ) using LR clonase ( Life Technologies ) . The Par1b-KD construct was described before [22] . HepG2 cells were cultured as previously described [23] . HepG2 cells expressing ABCB1-eGFP were cultured as previously described [24] . For experiments , cells were plated on ethanol-sterilized glass coverslips at a density of 5×104 cells/cm2 and grown for 2 d . For Par1b knockdown experiments , cells were plated at 15×104 cells/cm2 to match WIF-B9 conditions ( see below ) . WIF-B9 cells were grown in modified F-12 Coon's modification medium ( F6636; Sigma ) supplemented with 10−6 M hypoxanthine , 4×10−8 M aminopterin , 1 . 6×10−6 M thymidine , 5% ( v/v ) fetal bovine serum ( 100–106; Gemini ) , 1% glutamax ( Invitrogen ) , 0 . 5 µg/ml amphotericin , and 10 mM HEPES . For culture maintenance , cells were seeded in plastic dishes at 10×103 cells/cm2 and cultivated up to 4 d before replating . For experiments , differentiated cultures ( 10–12 d ) were plated on water-prewashed glass coverslips ( EMS ) in MatTek chambers at 15×104 cells/cm2 . Madin-Darby canine kidney ( MDCK ) cells were grown in DMEM without phenol red ( 17–205; Cellgro ) supplemented with 10% fetal bovine serum ( S11050; Atlanta Biologicals ) and 2 mM L-glutamine . Stable MDCK cell lines expressing gp135-GFP and Par1b were generated from T23-MDCK cells . MDCK-Par1b cells were prepared as previously described [22] . Cells were maintained at 37°C in a 5% CO2 ( HepG2 and MDCK cells ) or 7% CO2 ( WIF-B9 cells ) humidified atmosphere . Mouse liver tissues 48 h after partial hepatectomy ( formalin-fixed paraffin-embedded ) and mouse liver tissue from 23-d-old mice , collected 2 d after weaning ( prepared as near-native tissue slices as previously described [25] ) , were prepared as previously described [26] . Formalin-fixed paraffin-embedded rat liver tissue ( collected 2 d after weaning ) was a kind gift from C . Desdouets ( INSERM , France ) . For HepG2 cells , RNA interference was performed using the pLKO lentiviral knockdown system ( http://www . addgene . org/tools/protocols/plko/ ) . The target sequences used for Par1b and LGN are listed in Table S2 and were generated according to the pLKO protocol . Knockdown was verified by real-time PCR on a StepOnePlus system ( Applied Biosystems ) using the primers listed in Table S3 . A short hairpin RNA ( shRNA ) –resistant Par1b was created by introducing missense mutations into the shRNA target sequence ( AGCAAGAGAGGCACTTTA to AGTAAAAGGGGAACATTG ) using a Q5 Site-Directed Mutagenesis Kit ( E0554S; New England Biolabs ) . All constructs were verified by sequencing . RNA was isolated using Tri-Reagent from Sigma ( T9424 ) . RNA interference experiments in WIF-B9 and MDCK cells were performed as previously described [22] . Lentiviral particles were created using a second-generation system based on pCMVdR8 . 1 ( structural components ) and pVSV-G ( envelope protein ) . Briefly , 2 . 6×106 HEK293T cells were plated in a 10-cm dish . The next day , the cells were co-transfected with CaPO4-DNA complexes of pCMVdR8 . 1 , pVSV-G , and either pLKO . 1 or pLenti constructs for ∼16 h . Medium was refreshed , and after 24 and 48 h viral particles were harvested and filtered through a 0 . 45-µm PVDF membrane filter . Viral supernatants were stored at −80°C . 1-d-old HepG2 cells were infected with viral particles for 24 h , whereafter cells were incubated with normal growth medium to recover from the viral infection . Selection medium ( 1 µg/ml puromycin; Sigma ) was added 24 h later to select for transduced cells . WIF-B9 cells expressing DPPIV-TagRFP , GFP , Par1b-DN-GFP , pSUPER-GFP , or shRNA Par1b-GFP were obtained by adenovirus-mediated transduction [22] in Opti-MEM ( Invitrogen ) for 1 h with one plaque-forming unit/cell and 10–12 h expression at 37°C . Cells were fixed in 4% paraformaldehyde at 37°C ( MDCK and WIF-B9 cells ) or for 20 min at room temperature ( HepG2 cells ) . For staining microtubular structures in HepG2 cells ( tubulin , NuMA , and LGN ) , cells were pre-extracted in 0 . 5% Triton X-100 in PHEM buffer ( 1 min ) , washed once in PHEM buffer , and fixed in 4% paraformaldehyde in PHEM buffer . For HepG2 cells , blocking and permeabilization were performed for 30 min at room temperature in HBSS containing 0 . 025% saponin ( w/v ) , 1% ( w/v ) BSA , and 0 . 02% sodium azide , followed by antibody staining in the same buffer . WIF-B9 and MDCK cells were permeabilized with 0 . 2% Triton X-100 and blocked with 1% BSA . Antibody incubation was performed in PBS–1% BSA . HepG2 cells were imaged using a combination of widefield ( Olympus AX70 ) and confocal microscopy ( Leica SP2; HCX PL APO 63x/1 , 4 oil; pinhole 1 AU; pixel size 80 nm ) and analyzed using a combination of Imaris ( Bitplane ) , ImageJ , and Adobe Photoshop CS4 . A Solamere Nipkow confocal live cell imaging system was used ( HCX PL APO 63x/1 , 3 glycerin; pixel size 117 nm ) to live-image z-stacks of 7×1 . 5 µm every 4 min , unless otherwise indicated . WIF-B9/MDCK cells were imaged with a TCS SP5 confocal microscope ( Leica Microsystems ) equipped with a motorized x-y stage for multiple position finding and with an 8 , 000-Hz resonant scanner . Fixed cells were imaged using a HCX PL APO 63x/1 . 4-0 . 60 oil λBL CS objective on glass coverslips mounted in non-hardening , glycerol-based aqueous mounting medium . Confocal ( pinhole 1 AU; pixel size 80 . 02 nm ) xyz-stacks were taken from the monolayer . Live cell imaging was performed using a HCX PL APO 40x/1 . 25-0 . 75 oil CS objective on MatTek or CELLview chambers . xyzt-stack frames ( pinhole 2–3 AU; pixel size 100 . 1–252 . 8 nm ) were recorded . Image analysis was performed using LAS AF 2 . 3 . 1 and ImageJ 1 . 45 software . Brightness and contrast were adjusted according to the Journal of Cell Biology guidelines , without changing gamma settings . For calculating the orientation of the mitotic spindle in cell lines , a line was drawn from the center of the apical lumen through the center of the mitotic spindle ( PA ) . A second line was drawn through the spindle poles ( spindle axis [SA] ) . When no spindle pole staining was performed , it was assumed that the spindle poles were localized in a straight line perpendicular to the metaphase plate . The angle between these lines ( SA/PA ) was calculated with the ImageJ measure angle tool and plotted accordingly . To study the orientation of cell division in rat and mouse liver tissue , a line was drawn through both spindle poles of a dividing cell and extrapolated to determine the plasma membrane domain to which the spindle poles were oriented . The orientation of the spindle poles was scored as oriented towards ( 1 ) the bile canaliculus , ( 2 ) the apicolateral domain , or ( 3 ) the basolateral ( sinusoidal ) or common lateral membrane . Microsoft Excel was used for calculations , and Graphpad PRISM was used to generate graphs . Graphs represent mean ± standard deviation of three independent experiments , unless otherwise specified . Sample sizes ( n ) in graphs represent the total sample size . The statistical significance of differences was determined using Student's t-test ( two-tailed , unpaired , with equal variance ) unless otherwise specified .
We first analyzed the orientation of the mitotic spindle relative to the apical PA in vivo in mitotic mouse hepatocytes that were in metaphase or in telophase 48 h after hepatectomy . A line drawn through the mitotic spindles poles ( immunolabeled with antibodies against the microtubule-binding nuclear mitotic apparatus protein [NuMA] ) typically intersected the dipeptidyl peptidase 4 ( DPPIV ) –positive apical canalicular domains ( Figure 2A , arrowheads ) or its flanking regions , rather than the basolateral/sinusoidal domains ( Figure 2A , sinusoidal domains are indicated by “si” and dotted lines ) . These data are in agreement with earlier observations in proliferating rat hepatocytes following hepatectomy [27] , [28] . Quantification of confocal images of 61 mitotic hepatocytes from three mice 48 h after hepatectomy ( see Materials and Methods ) revealed that 85 . 1%±10 . 1% of the mitotic spindle axes intersected the apical bile canalicular or apicolateral domain ( Figure 2B ) . We also analyzed the orientation of the mitotic SA relative to the apical PA in hepatocytes in vivo in fixed liver tissue from young healthy mice ( Figure S1A ) and rats ( Figure 2C and 2D ) , which display a burst of cell division after weaning [26] , [29] . A line drawn through the NuMA-labeled mitotic spindle poles of rat hepatocytes that were in metaphase or in telophase typically intersected the DPPIV-positive apical bile canalicular plasma membrane domain or its flanking region ( Figure 2C ) . We named this flanking region the apicolateral domain , as it could be distinguished from the “common” lateral plasma membrane facing neighboring cells with which no apical lumen was ( yet ) formed ( Figures 2D and S1B; Movie S1 ) , and is a geometrically distinctive feature of cells with a hepatic polarity phenotype ( Figure 1A ) . The orientation of the mitotic spindle in hepatocytes in vivo correlated well with a restricted localization of the mitotic-spindle-orientating protein LGN [30]–[33] at and/or in close proximity to the zona occludens 1 ( ZO-1 ) –marked tight junctions at this apicolateral region flanking the apical/bile canalicular domain ( Figure 2D , green lines in the diagram ) . LGN was largely absent from the “common” lateral plasma membrane facing neighboring cells with which no apical lumen was shared and absent from basal/sinusoidal plasma membrane domains ( Figure 2D , orange and grey lines , respectively , in the diagram ) . These data demonstrate that in hepatocytes in vivo , the mitotic spindle is predominantly orientated towards an LGN-enriched apical/apicolateral surface domain . To investigate the dynamics of mitotic spindle orientation and cell division and its molecular regulation in living hepatocytes , we made use of the polarizing human hepatocyte cell line HepG2 [9] , [34] . HepG2 cells develop apical lumens amidst their lateral surfaces facing adjacent cells ( Figures 1A and 3 ) , reflecting the earliest stages of apical–basal polarity development in the fetal liver [5] , [7] , [8] , [23] , [35] . Importantly , in agreement with the observations in hepatocytes in vivo , a line drawn through the spindle poles in HepG2 cells in more than 70% of all cases intersected the apical or apicolateral domain ( Figures 3A and S2A ) , where LGN was almost exclusively localized ( Figure 3B ) . Note that LGN was not detected at the “common” lateral and sinusoidal plasma membrane domains ( Figure 3B , apicolateral domains are green and “common” lateral and sinusoidal domains are orange and grey , respectively , in the diagram ) . To more accurately determine the orientation of the mitotic spindle relative to the apical PA in these cells , we measured the angle between the SA ( the line drawn through the spindle poles; Figure 3C , dotted line ) and the PA ( the line drawn between the center of the immunolabeled apical domain and the center of the mitotic spindle; Figure 3C , solid line ) . 50 . 9%±7 . 0% , 32 . 1%±6 . 8% , and 17 . 0%±0 . 3% of the mitotic spindle axes in cells that were in metaphase displayed an SA/PA angle of 0–30° , 31–60° , and 61–90° , respectively , with statistically significant differences between the three categories ( Figure 3D and 3E ) . A similarly biased SA/PA angle distribution was observed in cells that were in later stages of mitosis , i . e . , anaphase or telophase ( Figure S2A–S2C ) , and live cell imaging in HepG2 cells showed that the apicolateral-directed orientation of the mitotic spindle was fixed early in mitosis and remained stable throughout the subsequent mitotic stages ( Figure S2D and S2E; Movie S2 ) . These data demonstrate that HepG2 cells , similar to hepatocytes in vivo , orient their mitotic spindle with a significant bias towards an LGN-enriched apicolateral plasma membrane domain . These cells are therefore a useful model system to study the consequences of this stereotypic mitotic spindle orientation with regard to cell polarity and its molecular regulation . Concomitant with the predominant apicolateral-plasma-membrane-directed orientation of the mitotic spindle apparatus , live cell imaging revealed that HepG2 cells predominantly divided in such a way that only one of the two emerging daughter cells inherited the apical lumen . Stills from a representative movie ( Movie S3 ) of dividing HepG2 cells that express the green fluorescent eGFP-tagged apical protein ABCB1 ( Figure 4A ) show two non-mitotic ( interphase ) cells with ABCB1-eGFP-positive apical plasma membrane domains and lumens ( indicated by the red arrowhead ) . After each cell passed through metaphase it formed a cleavage furrow ( Figure 4A , black arrowheads ) during anaphase/telophase that , following subsequent cytokinesis , gave rise to one daughter cell ( marked by “1” ) that inherited the apical domain ( red arrowhead ) and one daughter cell ( marked by “2” ) that did not . The vast majority of live cells ( >75% ) showed this asymmetric segregation of the apical plasma membrane domain and lumen during mitosis ( Figure 4B ) . Similar results were observed with another hepatocyte cell line , WIF-B9 ( Figure S3A and S3B; Movie S4 ) , underscoring that this mode of cell division orientation is a feature of polarized hepatocytes and not only of HepG2 cells . Interestingly , we observed that the emerging non-polarized daughter cells could reestablish apical–basal polarity and reestablish an apical lumen with their new neighbor cells . An example of this is shown in the stills ( Figure 4C ) from Movie S5 . These stills show that a HepG2 cell with an ABCB1-eGFP-positive apical plasma membrane domain and lumen ( red arrowhead ) formed a cleavage furrow ( white arrowheads ) that gave rise to one daughter cell ( marked by the asterisk ) that did not inherit the apical domain ( red arrowhead ) , but reestablished an ABCB1-eGFP-positive apical domain ( yellow arrowhead ) with its sister at the site of cytokinesis . Taken together , we conclude that hepatocytes orientate their mitotic SA with a significant bias towards an LGN-enriched apicolateral plasma membrane domain , and asymmetrically segregate their apical plasma membrane domain and apical lumen to the emerging daughter cells . In order to investigate to what extent the hepatocyte polarity phenotype—as such—was sufficient to dictate an apicolateral-domain-directed orientation of the mitotic spindle , we made use of our earlier observation that the overexpression of the polarity protein partitioning-defective 1/microtubule-affinity regulating kinase 2 ( Par1b/MARK2 ) in simple epithelial cells induces a hepatic polarity phenotype ( schematically depicted in Figure 5A ) [22] . Thus , when Par1b was overexpressed in MDCK cells ( a widely used model of simple epithelial cells ) , apical plasma membrane proteins such as gp135 localized to lumens formed between adjacent cells ( Figure 5B , Par1b ) , rather than to the cell-culture-medium-facing cell surface at the top of the control cell monolayer ( Figure 5B , control ) . In mitotic cells identified in parental MDCK cell cultures , LGN was localized at the lateral plasma membrane domains and was excluded from the apical domain ( Figure 5B , control , dotted white line; and previous reports [19] ) . In agreement with the role of cortical LGN in spindle orientation in these cells [19] , 100% of all SA/PA angles have been demonstrated to be in the 61–90° range , giving rise to the symmetric segregation of apical and basal domains to both daughter cells [36] , [37] . In contrast , in MDCK cells that overexpressed Par1b and displayed a hepatic polarity phenotype , the localization of LGN was highly polarized and restricted to the apical/apicolateral domain ( Figure 5B , Par1b , red arrowhead ) , and LGN was excluded from the “common” lateral cell surfaces ( Figure 5B , Par1b , dotted white line ) . Coinciding with this change in the distribution of LGN , 58 . 3%±7 . 5% , 32 . 4%±6 . 6% , and only 9 . 3%±5 . 8% of all SA/PA angles were in the 0–30° , 31–60° , and 61–90° range , respectively , with a clear , statistically significant difference between these categories ( Figure 5C ) . These data demonstrate that the overexpression of Par1b in simple epithelial cells caused a strong shift from a mitotic spindle orientation that was perpendicular to the apical PA to one that was significantly more parallel to the apical PA , matching a change in the distribution of LGN . Concomitantly , live cell imaging revealed that Par1b-overexpressing MDCK cells , like hepatic cells , predominantly divided in such a way that only one of the two emerging daughter cells inherited the apical lumen ( Figure S4A–S4C; Movie S6 ) . These data show that Par1b coordinates the induction of a hepatic polarity phenotype with a change in ( 1 ) the localization of LGN , ( 2 ) the orientation of the mitotic spindle , and ( 3 ) the asymmetric segregation of the apical plasma membrane domain to the emerging daughter cells . The results from the Par1b-overexpressing MDCK cells as displayed in Figure 5 do not demonstrate whether solely the induction of a hepatic polarity phenotype—as such—was sufficient to change the localization of LGN and the orientation of the mitotic spindle . Therefore , in order to further investigate to what extent Par1b is important for the localization of LGN and spindle orientation in the context of the hepatic polarity phenotype , we knocked down Par1b in HepG2 cells ( Figure S5A ) . Confocal images in Figure 6A and 6B show that in control cells—i . e . , cells treated with scrambled shRNA ( scramble ) —LGN ( black arrowheads ) localized at the cell surface flanking the ABCC2- or ZO-1/actin-labeled apical domain ( red arrowheads ) , similarly to in untreated cells ( cf . Figure 3B ) . In contrast , in cells treated with shRNA against Par1b ( Par1b-KD ) , the localization of LGN was no longer restricted to the apicolateral domain ( Figure 6A and 6B ) . LGN frequently showed an ( additional ) localization at the “common” lateral domain ( black arrowheads , the “common” lateral domain is orange in the diagram ) , away from the apical domain ( Figure 6A and 6B , red arrowhead ) . Note that these cells have retained the typical hepatic polarity phenotype . The change in the distribution profile of LGN was accompanied by a change in the orientation of NuMA-positive astral microtubules that emanated from the mitotic spindle poles and reached out to the cell cortex ( z-stack sections in Figure S5B; Movies S7 and S8 ) . The knockdown of LGN in HepG2 cells with two different shRNAs ( Figures 6C and S6A ) caused a randomization of the SA/PA angle in the 0–30° , 31–60° , and 61–90° categories , and caused a significant reduction of SA/PA angles in the 0–30° category when compared to control cells ( Figures 6D , S6B , and S6C ) , hence underscoring the contribution of LGN to mitotic spindle orientation in these cells . In agreement with the change in LGN distribution , the knockdown of Par1b caused a randomization of the orientation of the mitotic spindle axes , with approximately 26 . 8%±9 . 2% , 32 . 6%±10 . 1% , and 40 . 6%±19 . 2% displaying an SA/PA angle between 0–30° , 31–60° , and 61–90° , respectively , with no statistically significant difference between the three categories ( Figure 6E ) . Thus , Par1b knockdown effectively abolished the bias in mitotic SA orientation towards smaller SA/PA angles ( 0–30° ) . An illustrative example of this is shown in Figures 6A , 6B , and S5D , and the quantifications are depicted in Figures 6E and S5C . The reintroduction of shRNA-resistant Par1b ( Figure S5A ) completely rescued this effect , and treatment of the cells with a scrambled shRNA without effect on Par1b expression did not affect the SA/PA distributions ( Figure 6E , control and rescue ) . Similarly to in HepG2 cells , the knockdown of Par1b in WIF-B9 cells , or the expression of a dominant-negative Par1b mutant , caused a statistically significant shift in SA/PA angle bias from 0–30° to 31–60° and 61–90° ( Figure S7A–S7C ) , underscoring that the role of Par1b mitotic spindle orientation is a feature of polarized hepatocytes and not only of HepG2 cells . Concomitant with the loss of bias towards 0–30° angles , live cell imaging showed that the frequency of cell divisions in which both daughter cells inherited part of the same apical lumen significantly increased upon Par1b depletion or expression of the Par1b-DN mutant ( Figure S7D–S7F; Movie S9 ) . Taken all together , our data suggest that , in cells with a hepatic type polarity , Par1b controls the apicolateral enrichment of LGN and , thereby , the apicolateral-directed orientation of the mitotic spindle to promote the asymmetric segregation of the apical plasma membrane domain to the two emerging daughter cells and to preserve the typical polarity phenotype of polarized hepatocytes .
This study demonstrates that mitotic hepatocytes asymmetrically segregate their apical plasma membrane domains to the emerging daughter cells during cell division . This is in striking contrast to the symmetric segregation of apical and basal surface domains observed in vitro and in vivo in simple epithelial cells such as those found in the neuroepithelium , kidney , and intestine ( reviewed in [15] ) . Our data indicate that this asymmetric inheritance of the apical plasma membrane domain in hepatocytes is dictated by an apicolateral-plasma-membrane-domain-directed orientation of the mitotic spindle . Interestingly , this apicolateral plasma membrane domain is a geometrically distinctive feature in cells with a ( fetal ) hepatic polarity phenotype ( Figure 1A , apicolateral domain is green in the diagram ) . Indeed , the apicolateral domain represents the cell's contacting surface with the adjoining cell with which it shares an apical lumen , and can be distinguished from its contacting surface with other adjoining cells with which no apical lumens are ( yet ) shared ( Figure 7 , orange ) . This apicolateral subdomain has gone unnoticed , presumably because no functional relevance had been ascribed to it . Our data now demonstrate that Leu-Gly-Asn repeat-enriched protein ( LGN ) predominantly accumulates at this apicolateral domain during mitosis , both in rat liver hepatocytes in vivo and in polarized HepG2 cells in culture . Furthermore , NuMA-positive astral microtubules predominantly target this apicolateral domain in mitotic HepG2 cells . These observations are consistent with data from other cell systems in which LGN recruits NuMA on astral microtubules to the cell cortex ( reviewed in [13] , [15] , [38] ) . Indeed , knockdown experiments demonstrate that LGN is necessary for orientating the SA predominantly towards the apicolateral domain . We propose that this apicolateral domain thus serves as an instructive positional landmark in hepatocytes for the polarized recruitment of LGN , which , in turn , is required for the predominantly apicolateral orientation of the mitotic spindle and asymmetric segregation of the apical domain to the nascent daughter cells . In contrast to the apicolateral accumulation of LGN in dividing hepatocytes , LGN has been shown to accumulate at the apical plasma membrane domain in asymmetrically dividing Drosophila neuroblasts [39] , [40] . In epithelial cells , atypical protein kinase C at the apical plasma membrane has been proposed to exclude the apical recruitment of LGN [41] , [42] . Possibly , the presence of atypical protein kinase C at the apical bile canalicular plasma membrane domain in HepG2 cells and primary hepatocytes ( unpublished data ) may have similarly prevented the accumulation of LGN at the apical surface . But what caused LGN accumulation at the apicolateral subdomain and excluded it from the “common” lateral surface in hepatocytes ? Our data implicate the polarity protein Par1b as a critical determinant for this . Indeed , upon knockdown of Par1b in HepG2 cells , LGN no longer accumulated predominantly at the apicolateral domain , but rather showed additional localization at “common” lateral plasma membrane domains . In agreement with the occurrence of multiple sites of cortical LGN , NuMA-positive astral microtubules reached multiple sites at the cell surface . Concomitant with the altered distribution of LGN , knockdown of Par1b or expression of a nonfunctional Par1b mutant in HepG2 and WIF-B9 cells resulted in a loss of spindle orientation bias towards the apicolateral domain and promoted symmetric cell divisions that bisected the apical surface . Notably , in Par1b-depleted hepatic cells the loss of apicolaterally restricted LGN occurred while cells maintained their apicolateral domain . This suggests that Par1b translated the presence of the apicolateral domain as a positional landmark to cortical polarity—i . e . , the apicolateral accumulation of LGN—in the mitotic cell . This is further supported by our observations that the overexpression of Par1b in simple epithelial MDCK cells coordinated the acquisition of a fetal hepatocyte polarity phenotype ( and thus the establishment of an apicolateral domain ) with apicolateral LGN recruitment during mitosis , spindle orientation , and asymmetric cell division . This demonstrates that the coordinated action of Par1b and LGN constitutes a fundamental part of the molecular mechanism that drives mitotic spindle orientation and asymmetric/symmetric apical plasma membrane inheritance . Further studies are needed to determine how Par1b precisely controls the exclusive apicolateral recruitment of LGN . The orientation of cell division parallel to the apical–basal axis establishes cell fate specification , as has been shown in skin epithelial cells [43] , [44] and in neuroblasts [15] , although not necessarily [31] . Apart from the asymmetric inheritance of apical plasma membrane proteins , we observed no overt signs of cell fate specification in dividing HepG2 cells . Live cell imaging showed that emerging daughter cells that did not inherit the apical plasma membrane domain were capable of establishing an apical plasma membrane domain and lumen with a new neighbor or the sister cell , and did not show distinct behavior when compared to the polarized daughter cell . We cannot , however , exclude the asymmetric acquisition of specific molecules that may have endowed one of the daughter cells with distinct capabilities . Although fetal liver development/patterning has not been experimentally tested , our findings may suggest that in early fetal hepatocytes the asymmetric segregation of apical domains during division may , in conjunction with a repolarization of non-polarized nascent daughter cells , promote the dissemination of isolated apical lumens throughout the proliferating cell mass . It can be speculated that , in vivo , such asymmetric-cell-division-driven propagation of biliary luminal pockets throughout the proliferative fetal liver parenchymal mass could facilitate the development of a branching canalicular network via concomitant or subsequent apical lumen expansion and fusion ( Figure 7 ) . Indeed , elongation of bile canaliculi results from the expansion and fusions of numerous small isolated apical lumens [1] , [2] , [5] and as such would not necessarily require symmetric cell divisions . We propose that the Par1b-regulated spindle orientation via LGN and the resultant asymmetric inheritance of individual apical plasma membrane domains and lumens , as shown in this study , are made possible by and serve the unique polarized architecture of hepatocytes and , possibly , the liver parenchymal tissue . To our knowledge , there have been no reports in the literature on human liver diseases associated with Par1b , LGN , or hepatic spindle disorientation . Defects in the orientation of the mitotic spindle apparatus may hamper the efficient development of bile canalicular networks during normal liver development or regeneration , or promote the development of cystic lumens , the latter process typically being driven by symmetric divisions [17]–[19] . In future studies Par1b knockout mice may be useful to investigate the role of Par1b in the formation of the bile canalicular network during embryonic liver development or regeneration after hepatectomy .
|
The development and maintenance of the polarized epithelial architecture and function of organs that form tubular “lumen” structures is important for normal physiology and , when deregulated , gives rise to disease . Recent studies have highlighted the importance of a strict coordination of the orientation of mitotic divisions relative to an internal axis of asymmetry in proliferating epithelial cells during this process . Hepatocytes are the predominant epithelial cells of the liver . Hepatocytes display a unique lumen-forming architecture and cellular asymmetry , but the molecular basis for this special polarized architecture is not well understood . Our study now reveals an unexpected mode of plasma membrane domain inheritance that is coupled to a cellular axis of asymmetry in proliferating mammalian hepatocytes . We show that mitotic hepatocytes asymmetrically segregate their apical plasma membrane ( the membrane facing the lumen structure ) along with the lumen to their daughter cells . We demonstrate that the coordinated action of two proteins , Par1b and LGN , constitutes a fundamental part of the underlying molecular mechanism . This coupling of cell division and polarity in hepatocytes is distinct from that established in other epithelial cell types . These findings are important for understanding the unique polarized tissue architecture in the developing liver .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2013
|
Par1b Induces Asymmetric Inheritance of Plasma Membrane Domains via LGN-Dependent Mitotic Spindle Orientation in Proliferating Hepatocytes
|
Lipid rafts are membrane microdomains that function as platforms for signal transduction and membrane trafficking . Tyrosine kinase interacting protein ( Tip ) of T lymphotropic Herpesvirus saimiri ( HVS ) is targeted to lipid rafts in T cells and downregulates TCR and CD4 surface expression . Here , we report that the membrane-proximal amphipathic helix preceding Tip's transmembrane ( TM ) domain mediates lipid raft localization and membrane deformation . In turn , this motif directs Tip's lysosomal trafficking and selective TCR downregulation . The amphipathic helix binds to the negatively charged lipids and induces liposome tubulation , the TM domain mediates oligomerization , and cooperation of the membrane-proximal helix with the TM domain is sufficient for localization to lipid rafts and lysosomal compartments , especially the mutivesicular bodies . These findings suggest that the membrane-proximal amphipathic helix and TM domain provide HVS Tip with the unique ability to deform the cellular membranes in lipid rafts and to downregulate TCRs potentially through MVB formation .
Lipid rafts are membrane microdomains that take part in coordinating cell signalling functions and membrane trafficking . In T cells , upon antigenic stimulation , T cell receptors ( TCRs ) are recruited to lipid rafts , where they transmit signals via several pathways . The TCR signals induce the anchoring of lipid rafts to the underlying actin cytoskeleton , resulting in the assembly of lipid rafts [1] . Subsequently , clustered lipid rafts , containing TCR/CD3 complexes , are subjected to endocytosis , and the TCR/CD3 complexes are targeted for lysosomal degradation [2] . Thus , current evidence indicates that lipid rafts function as platforms for both the signalling and endocytosis of activated TCRs . Despite the important role of lipid rafts in signalling and membrane trafficking in T cells , the regulatory mechanisms controlling membrane trafficking to lysosomal compartments remain unclear . Several biochemically distinct compartments for membrane trafficking have been identified in other cell types including primary endocytic vesicles , early endosomes , late endosomes , and lysosomes . It has been recently demonstrated that multivesicular bodies ( MVB ) , also known as vesiculated late endosomes , are required for many key trafficking processes such as the downregulation of activated signalling receptors [3] . However , difficulties in elucidating the mechanisms of membrane trafficking have been compounded in T cells , because the fate of endocytic vesicles and the dynamics of transport intermediates remain uncertain . Herpesvirus persists in its host by entering a latent state , periodically reactivating to produce infectious viral particles . Herpesvirus saimiri ( HVS ) , an oncogenic γ2 herpesvirus , persists in the T lymphocytes of its natural host , the squirrel monkey , without any apparent disease symptoms , but infection of other species of New World and Old World primates results in fulminant T cell lymphomas [4] . In addition , when HVS infects the primary T lymphocytes of humans , Old World primates , New World primates , or rabbits , it can immortalize infected T cells , allowing them to grow independently of IL-2 [5] . Tyrosine kinase-interacting protein ( Tip ) is encoded in the first open reading frame at the left end of the highly oncogenic strains of HVS . Tip is not required for viral replication , but is required for T cell transformation in cultures , and for lymphoma induction in primates [4] . Tip has multiple binding sites for cellular proteins . The interaction of Tip with Lck kinase , which is mediated by the Src homology 3-binding ( SH3B ) motif and C-terminal Src-related kinase homology ( CSKH ) domain of Tip [6] , [7] , interferes with early events in the TCR signal transduction pathway , resulting in inhibition of immunological synapse formation [8] . Tip also interacts with p80 , a novel cellular endosomal protein that contains an N-terminal WD repeat domain and a C-terminal coiled-coil domain [9] . The interaction of Tip with p80 , which is mediated by a region containing a serine-rich ( SR ) motif , facilitates the formation of enlarged lysosomal vesicles , and results in the targeting of Lck and TCR/CD3 complexes for lysosomal degradation . We have previously demonstrated that Tip constitutively localizes in lipid rafts and exploits Lck and p80 to recruit TCR/CD3 complexes , leading to lipid raft aggregation and internalization [10] . Constitutive localization of Tip in lipid rafts depends on the C-terminal transmembrane ( TM ) domain , but not Lck and p80 interaction , and is also necessary for the efficient downregulation of TCR/CD3 and CD4 surface expression without affecting the inhibition of TCR signal transduction [11] . In this study , we report the presence of a putative amphipathic helical motif preceding the TM domain of Tip . Structural analysis revealed that Tip's amphipathic helical motif is composed of hydrophobic and positively-charged amino acid residues . Recently , the amphipathic helical motif has attracted much attention due to its active role in membrane curvature formation and membrane trafficking [12] . Thus , we investigated roles of the amphipathic helical motif in the molecular functions of Tip , including lipid raft localization and downregulation of TCR/CD3 and CD4 . We found that the membrane-proximal amphipathic helical motif is required for the efficient localization of Tip in lipid rafts as well as its selective downregulation of TCR/CD3 , potentially through deformation of membrane structures and MVB formation in T cells .
We have recently reported that the TM domain ( amino acid residues 229–250 ) of Tip is required for its association with lipid rafts , while other motifs involved in interactions with Lck and p80 are dispensable for lipid raft targeting [11] . In this study , GFP-Tip fusion proteins carrying deletions from the cytoplasmic region of Tip were generated ( Figure S1 ) , and the motif required for lipid raft localization was mapped . The degree to which Tip was associated with lipid rafts , in 293T cells , was estimated by densitometry and represented as the average percentage value from triplicate samples . The position and the integrity of lipid rafts in the discontinuous sucrose gradient were determined by the presence of GM1 ganglioside , which associates reliably with lipid rafts ( Figure 1A ) . We found that the wild type GFP-Tip fusion protein was efficiently associated with lipid rafts ( approximately 75% ) , which is consistent with our previous results [10] , [11] . Deletion of Tip's cytoplasmic domain , which contains known protein interaction motifs , had no discernible effect on the association of Tip with lipid rafts ( GFP-Tip184-256 ) . Two additional deletion mutants , GFP-Tip197-256 and GFP-Tip211-256 , showed a similar Tip distribution; approximately 50% was associated with lipid rafts . A GFP-Tip227-256 mutant carrying only the Tip TM domain was detected primarily in the fractions where cytoplasmic GFP protein localized , and showed only 20% association with lipid rafts . These results suggest that the C-terminal cytoplasmic regions proximal to the TM domain might contribute significantly to Tip's localization to lipid rafts . To exclude the potential confounding effects by GFP fusion on lipid raft association of Tip , we also constructed flag-tagged version of Tip mutants and examined their localization on lipid rafts ( Figure S2 ) . In this independent experiment , similar level of lipid raft association of flag-tagged Tip mutants was observed when compared with those of GFP-fusion proteins , indicating that GFP fusion does not significantly affect the lipid raft association of Tip and its mutants . We next analyzed the potential structure of the C-terminal regions spanning residues 184 to 256 of Tip using a protein structure prediction server [13] . The secondary structure analysis predicted with high confidence that the amino acid residues from 213 to 250 of Tip would form an α-helix ( Figure S3 ) . Notably , the amino acid residues from 213 to 228 , proximal to the TM domain , are composed of hydrophobic and positively-charged amino acid residues , and are predicted to form an amphipathic helical structure ( Figure 1B ) . This amphipathic helical motif was highly conserved in Tip from three different strains of HVS , and in Tio ( Two-in-one ) of Herpesvirus ateles ( HVA ) , a recently identified member of the γ2-herpesvirus family ( Figure 1C and Figure S3 ) . Tio , an oncoprotein of HVA , has been shown to induce transformation of T cells in a manner similar to that seen in StpC and Tip of the C488 strain of HVS [14] . These findings suggest that this potential amphipathic helical motif preceding the TM domain might play a role in Tip function . To evaluate the effect of the amphipathicity of Tip's membrane-proximal helical motif upon its lipid raft localization , this motif was mutated by replacing the hydrophobic and charged residues with lysine and alanine , respectively ( Figure 2A ) . The resulting ability of the Tip mutants to associate with lipid rafts was then assessed . When Tip's four conserved hydrophobic residues ( I216 , L220 , L223 , and I227 ) , predicted to form hydrophobic face , were replaced with lysines ( Tip amp1 ) , there was a ∼50% reduction in lipid raft association , in comparison to wild type Tip ( Figure 2B ) . The degree of lipid raft association observed in this mutant was similar to that observed in the GFP-Tip227-256 mutant , carrying only the TM domain of Tip ( Figure 1A ) . As such , these data suggest that the four conserved hydrophobic residues are critical for targeting of Tip to the lipid rafts . Sequential replacement of two or three consecutive , topologically-adjacent hydrophobic residues with lysine residues ( Tip amp1-2K . 1∼Tip amp1-3K . 2 ) resulted in a gradual reduction in Tip's lipid raft association , ranging from 16% to 54% of the degree of association observed in wild type Tip . Both Tip amp1-2K . 3 and Tip amp1-3K . 2 mutants carrying lysines proximal to the TM domain showed a low degree of lipid raft association , similar to that observed in Tip amp1 , indicating that the hydrophobic isoleucine and leucine residues proximal to the TM domain are more critical for the association with lipid rafts than are the more distal residues . Substitution of the positively-charged residues with alanine ( Tip amp2: R214 , K218 , K221 , and R222 ) also resulted in a ∼40% reduction in lipid raft association compared to wild type , demonstrating the significant contribution of these residues to Tip's localization to lipid rafts . Tip-mediated downregulation of TCR/CD3 and CD4 depends on its ability to associate with lipid rafts [11] . To examine the contribution of amphipathicity of Tip's membrane-proximal helix to this downregulation , levels of TCR , CD3 , CD4 , and CD45 surface expression were examined in Jurkat T cells stably expressing wild type Tip , Tip amp1 , or Tip amp2 , using flow cytometry . As shown previously [9] , expression of wild type Tip in T cells effectively downregulated the surface expression of TCR/CD3 and CD4 ( Figure 3A ) . In striking contrast , the downregulation of TCR and CD3 surface expression was severely impaired in Jurkat T cells expressing Tip amp1 or Tip amp2 ( Figure 3A ) . However , downregulation of CD4 was not significantly affected by the mutations which abolished the amphipathicity of the membrane-proximal helix . Neither wild type Tip nor its mutants had any significant effect upon the surface expression of CD45 , demonstrating the specificity of Tip's effects for TCR and CD3 downregulation . These results indicate that the amphipathicity of Tip's membrane-proximal helix is involved in the downregulation of TCR/CD3 , but not CD4 surface expression . We have previously shown that Tip's targeting to the lysosomal compartments involves its formation of a complex containing Lck and p80 [9] . Tip's formation of this complex is correlated with its lipid raft association and the lysosomal degradation of TCR/CD3 complexes [10] , [11] . To examine whether loss of amphipathicity in Tip's membrane-proximal helix might affect the lysosomal localization of the viral proteins and TCR/CD3 , Jurkat T cells transiently expressing Tip or Tip amp1 were reacted with antibodies specific to EEA1 , an early endosomal marker , LAMP2 , a late endosomal/lysosomal marker , or CD3ζ and then examined under a confocal microscope ( Figure 3B ) . To quantitatively compare the degree of colocalization of the proteins in the vesicular compartments , we measured the Pearson correlation coefficient ( R ) values ( see Materials and Methods ) for each set of colocalizing proteins in 10 to 20 cells ( Figure S4 ) . Vesicles containing wild type Tip were weakly colocalized with EEA1 ( average R value = 0 . 16 ) , but were strongly colocalized with LAMP2 ( R = 0 . 76 ) or CD3ζ ( R = 0 . 82 ) , as shown previously [10] . In contrast , Tip amp1 displayed partial colocalization with EEA1 ( R = 0 . 51 ) and CD3ζ ( R = 0 . 60 ) but did not colocalize with LAMP2 ( R = 0 . 02 ) , indicating that the amphipathicity of Tip's membrane-proximal helix is required for efficient lysosomal targeting of Tip and TCR/CD3 . We have previously shown that Tip's TM domain is required for its lysosomal trafficking [11] . To investigate the role of the membrane-proximal amphipathic helix and TM domain in Tip's lysosomal localization , Jurkat T cells expressing GFP-Tip211-256 , GFP-Tip amp1211-256 , or GFP-Tip CD71TM211-256 were reacted with antibodies specific to EEA1 or LAMP2 ( Figure 4A ) . Colocalization of the fusion proteins with the endocytic markers were evaluated quantitatively for Pearson correlation coefficient ( Figure S5 ) . The intracellular vesicles containing GFP-Tip211-256 colocalized strongly with LAMP2 ( R = 0 . 82 ) , but weakly with EEA1 ( R = 0 . 21 ) , suggesting that the amphipathic helix and TM domain are sufficient for Tip fusion proteins to be delivered into the late endosomes or lysosomes . In fact , almost all of the intracellular vesicles containing GFP-Tip211-256 were costained with LAMP2 in the transfected T cells ( data not shown ) . Unlike GFP-Tip211-256 , both GFP-Tip amp1211-256 and GFP-Tip CD71TM211-256 , a Tip mutant carrying the TM domain of CD71 in place of that of Tip ( Figure S1 ) , were only partially localized in LAMP2-positive vesicles , showing slight preferential colocalization with EEA1 ( R values range from 0 . 34 to 0 . 46 , Figure 4A and S5A ) . These results were further confirmed in HeLa cells expressing these fusion proteins ( Figure 4B and S5B ) . The colocalization of lysosomes with vesicles containing GFP-Tip211-256 ( R = 0 . 54 ) was also more prominent than with vesicles containing GFP-Tip amp1211-256 ( R = 0 . 30 ) or GFP-TipCD71TM211-256 ( R = 0 . 37 ) , in HeLa cells . To determine in more detail the distribution of the GFP fusion proteins , Jurkat T cells expressing GFP-Tip211-256 or GFP-Tip amp1211-256 were analyzed by immunoelectron microscopy after staining with gold-conjugated anti-GFP antibodies . The gold signal was generally associated with intracellular vesicular compartments . Interestingly , GFP-Tip211-256 was frequently detected in luminal buddings of vesicular membranes , or in membranous complexes within the lumen ( Figure 5 ) , reminiscent of the process of MVB formation [15] . MVBs form by budding into the lumen of the vacuolar endosomes , which carry membrane proteins selected for the late endosomal route . They are thought to fuse with late endosomes or , following maturation , directly with lysosomes [15] . In cells expressing GFP-Tip amp1211-256 , GFP was generally associated with smaller vesicular compartments , most likely the early endosomes , as shown in Figure 4 . Associations of membrane curvature or MVBs with the fusion proteins were barely detectable ( Figure 5 ) . Taken together , it appears that the peptide encompassing the membrane-proximal helix and the TM domain of Tip might be involved in MVB formation in late endosomal compartments where Tip and its complex are degraded . Membrane curvature is an active means for creating membrane domains and organizing trafficking [12] . Several mechanisms have been suggested to constitute active cellular processes for the formation of membrane curvature , and these include changes in lipid composition , oligomerization of curvature scaffolding proteins , and the insertion of amphipathic helices into the lipid bilayer [12] , [16] . The possibility that the membrane-proximal amphipathic helix of Tip interacts with lipids was examined using a lipid binding assay . As shown in Figure 6A , a synthetic peptide derived from the membrane-proximal amphipathic helix of Tip ( Tip wt211-228 ) was found to bind to a series of negatively charged lipids including phosphatidic acid ( PA ) , phosphatidylserine ( PS ) , phosphatidylglycerol ( PG ) , cardiolipin , phosphatidylinositide ( PtdIns ) , and sulfatide , but did not bind to other neutral or positively-charged lipids such as triglyceride ( TG ) , diacylglycerol ( DAG ) , phosphatidylenthanolamine ( PE ) , phosphatidylcholine ( PC ) , cholesterol , or sphingomyelin . The binding specificity of the peptide to these lipids was further examined by probing an array of several lipids immobilized on nitrocellulose membranes at concentrations ranging from 100 pmol to 6 . 2 pmol . As demonstrated in Figure 6B , both wild type ( Tip wt211-228 ) and mutant ( Tip amp1211-228 ) peptides , carrying lysine residues instead of hydrophobic amino acids , were able to bind dose-dependently to PA , PS , and PG , with binding saturation occurring at approximately 50 pmol of lipids ( Figure 6B and Figure S6 ) . Interestingly , the mutant peptides derived from Tip amp1211-228 were also capable of binding to PE . The peptide-lipid interactions were further validated in a liposome binding assay , in which liposomes composed of 65% PC , 25% PS and 10% cholesterol were reacted with biotin-conjugated peptides , then cosedimented by ultracentrifugation . The coprecipitated peptides were resolved by gel electrophoresis and were subsequently probed with streptavidin-HRP conjugates . As shown in Figure 6C , approximately 50% of Tip wt211-228 peptides were cosedimented with liposomes , whereas less than 5% precipitated in the absence of liposomes . The Tip amp1211-228 peptides also precipitated after incubation with liposomes , even more efficiently than did the amphipathic wild type peptide . These data suggest that the hydrophobic residues of the amphipathic helix had little effect upon peptide-lipid binding properties . It remains a possibility , however , that these residues might restrict the affinity and preference of Tip for specific lipids . The influence of the amphipathic helix upon membrane curvature formation was examined using a liposome-based membrane deformation assay [17] , [18] , [19] . The peptides were incubated with liposomes , which have the same lipid composition as those used in Figure 6 , and subsequently examined by electron microscopy ( Figure 7 ) . Tip wt211-228 peptides resulted in efficient and robust formation of tubules with diameters of 20–40 nm , whereas Tip amp1211-228 did not . A laser light scattering assay revealed that incubation with Tip wt211-228 peptides dramatically altered the size distribution of the liposomes into ranges of 40–250 nm , whereas no significant changes in liposome size were detected following incubation with Tip amp1211-228 ( Figure 7 and Figure S7 ) . Collectively , the reported results indicate that the positively-charged residues within Tip's amphipathic helix confer a specific affinity for negatively-charged phospholipids , potentially through ionic interactions . The conserved hydrophobic residues enable the amphipathic helix to act like a wedge inserted into the membrane , to induce membrane curvature . Previously , we demonstrated that Tip can induce the aggregation of lipid rafts and enhance the recruitment of lipid raft-resident proteins , eventually forming large vesicular compartments in T cells [9] , [11] . These results suggested that Tip might oligomerize within membrane microdomains , inducing structural changes in the lipid bilayer . As such , the possibility of Tip oligomerization was investigated by coexpressing flag-tagged wild type Tip and GFP-Tip fusion proteins , then immunoprecipitating with an anti-flag antibody ( Figure 8A ) . Immunoblotting with an anti-GFP antibody revealed that the flag-tagged Tip co-precipitated with the GFP-Tip fusion protein , but not with GFP , suggesting that Tip interacts with itself . To determine the region responsible for Tip oligomerization , GFP-Tip mutants were included in the immunoprecipitation assay [11] . GFP-Tip mutants no longer binding with Lck ( TipmLBD ) or p80 ( TipΔ2 ) formed an immune complex with flag-tagged wild type Tip , indicating that Tip's Lck- and p80-binding motifs do not participate in Tip oligomerization . However , a Tip mutant carrying the TM domain of CD71 ( Tip CD71TM ) in place of its native one failed to interact with wild type Tip , suggesting that Tip's TM domain mediates its oligomerization . Recently , Mitchell et al . , reported that Tip is present as monomeric form in solution based on hydrogen-exchange mass spectrometry and circular dichroism study [20] . However , the recombinant protein they used contains only cytoplasmic region without transmembrane domain of Tip . Thus , our current result using the full-length Tip protein including transmembrane domain is more appropriate in reflecting natural status of Tip in vivo . Oligomerization of Tip was further confirmed using blue native polyacrylamide gel electrophoresis with detergent-solubilized 293T cells expressing GFP-Tip211-256 or GFP-Tip CD71TM211-256 . Truncated forms of the GFP fusion proteins , with short cytoplasmic domains , were used so as to minimize potential interactions with other cellular proteins , and to facilitate more accurate estimates of the size of the oligomeric protein . As shown in Figure 8B , a GFP fusion protein carrying the TM domain of wild type Tip ( GFP-Tip211-256 ) migrated as a ∼150 kDa protein , whereas GFP-Tip CD71TM211-256 migrated as a ∼40 kD protein , corresponding to the size of the monomeric form of GFP-Tip CD71TM211-256 . The size of the ∼150 kD protein complex is suggestive of homo-oligomers of four GFP-Tip211-256 monomers .
Lipid rafts contain proteins that retain their association with membrane lipids . These proteins are mostly GPI-anchored or acylated , but a few are transmembrane proteins , which are targeted to lipid rafts through their TM domain or through membrane-proximal determinants [21] . Here , we have found that the presence of a membrane-proximal amphipathic helix , located in Tip's cytoplasmic face , significantly contributed to Tip's localization in the lipid raft . Extensive mutagenesis analysis revealed that the residues forming both the hydrophobic ridge and the positively-charged face of the helical motif are important for Tip's efficient association with lipid rafts ( Figure 2 ) . The segregation of hydrophobic and polar residues into two opposite faces of the helical structure matches well with the chemistry of the membrane interface , and has been suggested to contribute to membrane adsorption [22] , [23] . It has also been suggested that the amphipathic helical motif might target caveolin to lipid rafts through partial insertion of the hydrophobic ridge into lipid bilayer , and electrostatic interaction of the charged surface with phospholipids [24] . Another lipid raft-residing protein , α-synuclein [25] , is also anchored to membranes by an elongated amphipathic helical structure [26] . Although the specificity of the amphipathic helical motifs for lipid rafts has been poorly defined , binding of α-synuclein to raft-like liposomes was shown to require acidic phospholipids , with a preference for phosphatidylserine [25] . In T cells , cholesterol and negatively-charged phospholipids are concentrated in the ordered raft domains upon antigenic stimulation [1] . Interestingly , a recent report showed that many signaling and transport proteins contain clusters of positively-charged amino acids , suggesting that those clusters could mediate the plasma membrane-targeting of proteins , through interaction with acidic phospholipids [27] . In this study , we showed that the amphipathic helical peptide of Tip specifically interacts with negatively-charged lipids such as PS , PA , PG , and PtdIns rather than neutral or amino phospholipids ( Figure 6 ) , suggesting that positively-charged amino acids within the amphipathic helical motif might associate with these negatively-charged lipids . This possibility is consistent with previous findings , which showed that the cytoplasmic leaflet of biological membranes is enriched in negatively-charged lipids , and that lipid rafts are enriched with PS , PA , and PG [28] . As suggested by the results of binding assays utilizing artificial liposomes containing cholesterol ( Figure 6C ) , however , the presence of the hydrophobic ridge might restrict the interaction of the peptide with lipid rafts . In fact , α-synuclein binds more strongly to membranes containing low or no cholesterol [25] and the binding affinity of an amphipathic peptide to unilamellar vesicles is reduced by the presence of cholesterol [29] . The rigidifying effect of cholesterol on phospholipid acyl chains may limit the penetration of the peptide into the bilayer interior . Taken all together , the interaction of the amphipathic helical region of Tip with lipid bilayer might be heavily dependant on the membrane lipid composition . Previously , we found that the TM domain of Tip is required for association with lipid rafts [11] . In this study , we found that the TM domain alone confers weak ( ∼20% ) lipid raft association ( Figure 1A ) but is sufficient to mediate oligomerization of Tip ( Figure 8 ) . These results suggested that the TM domain might play a cooperative role in lipid raft association together with the lipid binding amphipathic helix; i . e . the both domains are required for the efficient association of Tip with lipid rafts . With regard to the relationship between lipid raft association and protein oligomerization , varying results have been reported . In some cases , self-assembling or ligand-induced oligomerization was required for proteins to associate efficiently with the lipid raft [30] , [31] , whereas oligomerization and lipid raft formation were independent in other cases [32] . Alanine scan mutagenesis was performed in an effort to elucidate the role of the Tip TM domain in lipid raft localization and oligomerization ( Figure S8 ) . A lipid raft fractionation assay using Tip mutants carrying four consecutive alanine residues in the TM domain showed that two mutants , carrying alanine residues in the regions 240–244 or 249–253 , associate with lipid rafts to the same degree as wild type Tip , and that other mutants were able to associate with lipid rafts to a limited degree of approximately 40 to 47% ( Figure S8B ) . These observations imply that the amino acids 231–241 and the TVLS motif , which are thought to interact with the inner and outer leaflets of biological membranes respectively , were both required for efficient lipid raft association , suggesting that the specific amino acid sequences comprising the Tip TM domain might contribute to the interaction with lipid raft domains . The immunoprecipitation assay , using cells expressing AU1-tagged wild type Tip and flag-tagged Tip mutants generated by the alanine scan mutagenesis , showed that all mutants tested co-precipitated with wild type Tip ( Figure S8C ) . As such , oligomerization might be mediated by regions longer than four consecutive amino acids , or by multiple contacts within the TM domain , rather than by a single or local contact . Since no motif specific for the oligomerization of Tip could be identified , the precise relationship between protein oligomerization and lipid raft association has yet to be determined , and will be the topic of future studies . It could be questioned whether the mutations in the amphipathic helix or transmembrane domain of Tip could affect the efficiency of membrane association itself rather than lipid raft localization . However , when we examined the membrane association of Tip and its mutants , Tip amp1 or Tip CD71TM , without detergent treatment during membrane fractionation , there was no significant difference in the degree of membrane association of the Tip proteins ( Figure S9 ) . Rather , the degree of membrane association of Tip amp1 or Tip CD71TM was slightly enhanced when compared to that of Tip wt . This result suggested that the mutations in the amphipathic helix or transmembrane domain could change the raft localization property of Tip without significantly affecting the efficiency of membrane association itself . Vesicle trafficking involves dynamic remodeling of cellular membranes , for which the formation of local membrane curvature is a critical step [12] , [33] . It has recently been shown that generation of membrane curvature can be driven by the interplay between lipids and proteins , through several mechanisms [12] . An emerging theme among these mechanisms is the involvement of amphipathic peptides that partially penetrate the lipid bilayer , acting as wedges . Active insertion of helical peptides into the bilayer results in an increase of surface area in one leaflet , possibly generating spontaneous curvature in the bilayer . This local curvature is subsequently sensed and stabilized by other domains of the curvature-forming proteins , or by coat proteins . For example , the N-terminal amphipathic helix found in the BAR domain of amphiphysin and endophilin has been shown to cause local membrane curvature , which is stabilized by a banana-shaped lipid-binding domain [19] , [34] , [35] . Helical domains found in other proteins such as epsin , Arf , and Sar1 were also shown to generate local membrane curvature in an induced manner and subsequently recruit coat proteins to stabilize the curvature [17] , [18] , [36] . The power of an amphipathic peptide to generate membrane deformation was previously demonstrated when a designed 18-mer peptide was shown to form extensive , 40–50 nm diameter tubules from liposomes [37] . They showed that that the deformation of liposomes depended on lipid composition and peptide properties such as length and the ratio of hydrophobic to hydrophilic amino acids . In the present study , we showed that an amphipathic 18-mer peptide derived from Tip's membrane-proximal helix can efficiently induce membrane deformation in an in vitro liposome tubulation assay ( Figure 7 ) . The point mutation of the conserved hydrophobic residues in this peptide into basic lysine residues improved liposome binding , in comparison to the wild type peptide , but abolished tubulation ( Figure 7 ) . These results suggest that Tip's membrane-proximal amphipathic helix is likely to alter membrane structure in a manner similar to that employed by other cellular proteins containing amphipathic helices . In fact , GFP fusion peptides encompassing the helical region and TM domain of Tip were detected in vesicular curvatures and multivesicular structures within endocytic vesicles of Jurkat T cells ( Figure 5 ) , whereas the mutant fusion peptides were not . Luminal budding of the limiting membrane and the formation of MVBs in the late endosomal pathway are efficient mechanisms for targeting membrane proteins/receptors to lysosomes for degradation [3] . Through its amphipathic helical motif , Tip might initiate the luminal budding step , which might be further enhanced by oligomerization through the TM domain . Considering that Tip's structural influence on the lipid bilayer is generated in the cytoplasmic face of endocytic vesicles , and that Tip lacks a curvature-sensing/stabilizing domain , the luminal budding would likely be assisted by other cellular proteins . Recently , it was shown that an inverse BAR domain-like mechanism in the proteins IRSp53 and MIM ( missing-in-metastasis ) induces a membrane curvature opposite to that of BAR domains , and deforms membranes by binding to their interior , resulting in plasma membrane protrusions rather than invaginations [38] . Thus , Tip-associated luminal budding may be facilitated by cellular proteins with an inverse BAR domain-like mechanism , which may be recruited through direct or indirect interactions with Tip . Vps ( vaculolar protein sorting ) proteins have been shown to be involved in lysosomal degradation of activated receptors through the MVB-sorting pathway in yeast and mammals [3] , [39] , [40] . Some mutations in those genes resulted in an enlarged late endosomal compartment , presumably because of an inability to invaginate the limiting membrane to form the MVB . It would be interesting to identify whether the protein sorting machineries could mediate inverse BAR domain-like functions in Tip associated-MVB formation . Expression of Tip in T cells was previously reported to induce clustering of lipid raft domains as well as redistribution of TCR/CD3 complexes into lipid raft domains [9] , [10] . Tip expression can also reorganize raft domains and enhance the recruitment of raft-resident components [11] . Similarly , T cell activation leads to the segregation of plasma membrane domains to form TCR signaling clusters , and this is accompanied by the condensation of the plasma membrane , driven by activation-induced protein-protein interactions such as anchorage to the cytoskeleton [41] . The clustered raft domain platforms are subsequently internalized and degraded in the lysosome to attenuate TCR signaling [42] . Previously , we showed that the TM domain is essential for the downregulation of TCR/CD3 complexes and CD4 by Tip [11] . Downregulation of the membrane proteins , however , is mediated through different mechanisms [9] , [10] . Downregulation of TCR/CD3 complexes by Tip is dependent on its interaction with and kinase activity of Lck as well as the interaction with p80 , whereas downregulation of CD4 by Tip is dependent on the physical association with Lck only . In the present study , we showed that Tip's membrane-proximal amphipathic helix , consisting of 14 amino acids , was essential for the selective downregulation of TCR/CD3 complexes but not for CD4 . The effect of this short motif on the receptor trafficking might be linked to the functional properties of raft-targeting and membrane deformation , as mentioned above . Lysosomal targeting through MVB formation by the amphipathic helix and TM domain of Tip suggested how clustered TCR/CD3 complexes in lipid raft domains are targeted for lysosomal degradation in Tip-expressing T cells . In contrast to TCR/CD3 complexes , CD4 surface expression is downmodulated consistently by the Tip mutants lacking amphipathicity in their membrane proximal helix ( Figure 3 ) . As such , downregulation of CD4 surface expression could be mediated by different mechanisms . Although the molecular mechanisms of CD4 trafficking in resting and activated T cells are largely unknown , it is interesting to note that human immunodeficiency virus has dual arms , Nef and Vpu , to downregulate CD4 surface expression through distinct mechanisms [43] . Nef links mature CD4 to components of clathrin-dependent trafficking pathways at the plasma membrane , and perhaps also in intracellular compartments , leading to internalization and delivery of CD4 to lysosomes for degradation . Vpu , on the other hand , interacts with newly-synthesized CD4 in the endoplasmic reticulum , linking CD4 to the SCF ubiquitin ligase , and facilitating the entry of CD4 into the endoplasmic reticulum-associated degradation pathway . The Tip-associated molecular mechanisms controlling CD4 expression remain to be elucidated . The phenotypic resemblance of Tip and TCR activation , leading to Lck activation , recruitment of TCRs to lipid rafts and finally to lysosomal degradation , suggests that HVS Tip may pirate cellular signaling molecules to emulate TCR stimulation for viral persistence and pathogenesis . Epstein-Barr virus LMP2A , a functional homologue of HVS Tip , has also been shown to be mimic the B cell receptor signal transduction to maintain viral latency , allowing long-term survival of infected B cell [44] , [45] . This viral protein interacts with B-cell signaling proteins , such as Lyn and Syk , through its N-terminal cytoplasmic tail . LMP2A functions in lipid rafts to block translocation of the B-cell receptor into lipid rafts , which leads to inhibition of the subsequent signaling and accelerated internalization of the BCR-cell receptor upon stimulation . Thus , the study of HVS Tip may provide valuable insight into the conserved mechanisms employed by other γ-herpesvirus signal modulators to regulate lymphocyte functions and may have significant implications for the understanding of viral persistence and pathogenesis . In summary , we have shown that a potential membrane-proximal amphipathic helix preceding the TM domain of Tip is essential for efficient lipid raft localization and selective downregulation of TCR/CD3 , most likely through mechanisms involving membrane curvature and MVB formation in endocytic vesicles . Moreover , we could dissect the functional roles of the amphipathic helix and the TM domain in membrane deformation and oligomerization , respectively . These novel mechanisms of the viral protein could provide valuable insights into the functional relationship between lipid rafts and MVB formation and the molecular details of membrane trafficking of the key receptors in T cells .
Jurkat T cells were grown in RPMI , and 293T and HeLa cells were maintained in DME medium , supplemented with 10% FBS . Jurkat T cells were electroporated using a Bio-Rad electroporator at 260V and 975µF in serum-free RPMI medium . Lipofectamine2000 ( Invitrogen ) or calcium phosphate ( Clontech ) was used to induce transient expression of Tip in HeLa and 293T cells . Stable Jurkat T cell lines expressing Tip or its mutants were selected and maintained in the presence of puromycin ( 5 µg/ml ) . We tested the expression level of Tip or its mutants in each cell line by semi-quantitative RT-PCR using β-actin gene as internal control [46]and confirmed the similar level of expression in all the established cell lines ( data not shown ) . An anti-GFP antibody ( Santa Cruz Biotechnology ) , a CTB-HRP conjugate ( Sigma ) , an anti-EEA1 antibody , and an anti-LAMP2 antibody ( BD Bioscience ) were used for immunoassays . DNA fragments encoding Tip and its mutants were cloned into pFJ , pBabe or p3xFlag_CMV vectors ( Sigma ) using methods described previously [11] . GFP fusion proteins containing Tip or its mutants were made using pEGFP-C2 plasmids ( Clontech ) . PCR-based mutagenesis was performed to create the Tip mutants , using sequences described previously [10] , [11] . A peptide corresponding to the amphipathic helical region of Tip ( Tip wt211-228 , ANERNIVKDLKRLENKIN ) and a mutant peptide in which hydrophobic residues were replaced with lysines ( underlined; Tip amp1211-228 , ANERNKVKDKKRKENKKN ) were synthesized by Peptron Inc . A lysine residue was added to the C-terminus of each peptide to allow them to be conjugated with biotin . Lipid rafts were isolated using a method involving flotation on discontinuous sucrose gradients [10] . Briefly , 5×107 293T cells were washed with ice-cold PBS and lysed for 30 min on ice in 1% Triton X-100 in TNEV buffer ( 10 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 5 mM EDTA ) containing phosphatase inhibitors and protease inhibitor cocktail ( Roche ) . The lysates were further homogenized in a Wheaton loose-fitting Dounce homogenizer . Nuclei and cellular debris were pelleted by centrifugation at 900×g for 10 min . For the discontinuous sucrose gradient , 0 . 5 ml of cleared cell lysates were mixed with 0 . 5 ml of 85% sucrose in TNEV and transferred to a Beckman 14×89 mm centrifuge tube . Diluted lysates were overlaid with 4 ml of 35% sucrose in TNEV and finally 1 ml 5% sucrose in TNEV . Samples were then centrifuged in an SW41 rotor at 200 , 000×g for 20 h at 4°C , and 0 . 5 ml fractions were collected from the top of the gradient . Membrane-enriched fraction was prepared to examine the efficiency of membrane association of Tip and its mutants as described elsewhere with slight modification [47] . In brief , 293 T cells expressing GFP fusion proteins containing Tip or its mutants were harvested and resuspended in lysis buffer ( 50 mM Tris , pH 7 . 8 , 250 mM Sucrose , and 2 mM EDTA ) with protease inhibitor cocktail ( Roche ) . After incubation on ice for 10 min , cells were lysed by 30 strokes using Dounce homogenizer at 4°C . Cellular debris and nuclei were removed by centrifugation at 1000×g for 10 min at 4°C . The postnuclear supernatant was layered onto a 60% sucrose cushion and centrifuged at 160 , 000×g for 1 h at 4°C . The membrane fraction on top of the sucrose cushion was collected , diluted 1∶2 with cold phosphate-buffered saline ( PBS; 100 mM phosphate , 150 mM NaCl , pH 7 . 2 ) and pelleted at 100 , 000×g for 1 h at 4°C . The supernatant was discarded and the membrane pellet was rinsed twice with cold PBS , and pelleted at 20 , 000×g for 30 min at 4°C . The enriched membrane fraction was further used for SDS-PAGE and subsequent immunoblot assay . Blue Native PAGE was performed as described previously [48] with slight modifications . Homogenized cells were solubilized by adding Triton X-100 to a final concentration of 2 . 5% . After removing cellular debris by centrifugation , the whole-cell lysates were collected and resolved by native gel ( 10% ) electrophoresis . Resolved proteins were transferred to a PVDF membrane and detected by immunoblot assay . Aldorase from rabbit muscle ( ∼160 kDa , Sigma ) and bovine serum albumin ( monomer: ∼66 kDa , dimer ∼132 kDa , Sigma ) were used as molecular weight markers . Cells ( 5×105 ) were washed with RPMI medium containing 10% fetal calf serum , and incubated with fluorescein isothiocyanate-conjugated or phycoerythrin-conjugated monoclonal antibodies for 30 min at 4°C . After washing , each sample was fixed with 4% paraformaldehyde solution and flow cytometric analysis was performed with a FACScan ( Becton Dickinson Co . ) . Antibodies against CD3 ( SK7 ) , CD4 ( Leu-3a ) , CD45 ( HI30 ) , and αβTCR were purchased from BD Pharmingen . Cells were fixed with 4% paraformaldehyde for 15 min , permeabilized with 0 . 2% Triton X-100 for 15 min , and reacted with primary antibodies in PBS for 30 min at room temperature . Alexa 488- or Alexa 594-conjugated anti-rabbit or anti-mouse antibodies ( Molecular Probes ) were used as secondary antibodies . Confocal microscopy was performed using an Olympus FV1000 laser-scanning microscope ( Olympus ) fitted with a 60× Olympus objective . Images were collected at 512×512 pixel resolution using Olympus imaging software . The stained cells were optically sectioned in the z-axis , and the images in the different channels ( photo multiplier tubes ) were collected sequentially . The images were rendered using Olympus Fluoview v1 . 6b or Adobe Photoshop software . To quantify the degree of relative colocalization , we obtained the Pearson correlation coefficient ( R ) values , which are standard measures of colocalization [49] . The R values were calculated using the Olympus Fluoview v1 . 6b colocalization module which generates a “colocalized” image from two channels . For immunoprecipitation , cells were harvested and resuspended in lysis buffer ( 150 mM NaCl , 0 . 5% Nonidet P-40 , and 50 mM HEPES buffer , pH 7 . 4 ) containing protease inhibitors . Immunoprecipitated proteins from precleared cell lysates were used for immunoblot . For immunoblot , polypeptides were resolved by SDS-PAGE and transferred to a PVDF membrane . Immunoblot detection was performed with a 1∶1000 or 1∶3000 dilution of primary antibody and an enhanced chemiluminescence system ( Pierce ) . Membrane lipid strips and arrays ( Echelon Biosciences ) were used for peptide-lipid binding assays according to the manufacturer's instructions . Peptides ( 0 . 4 µM ) were incubated overnight at 4°C and detected with streptavidin-HRP conjugates . Densitometric analysis was applied to determine the relative affinity of peptide binding to the various lipids . After subtracting background values , numerical densitometric values were attributed to each of the five concentrations measured . The highest value , for binding of peptides to 100 pmol of lipids , was arbitrarily assigned “100% binding” and all other lipids were normalized in comparison to that maximum binding value . Synthetic liposomes were made using phosphatidylcholine ( 65% mol/mol ) , phosphatidylserine ( 25% mol/mol ) , and cholesterol ( 10% mol/mol; Avanti Polar Lipids Inc . ) , as described previously [19] . To achieve desired diameters , the liposomes were extruded more than 10 times through a polycarbonate membrane ( Avanti ) . The size of the liposomes was measured by laser light scattering analysis ( Brookhaven Instruments Co . ) . For the peptide binding assays , liposomes were diluted in 100 µl of binding buffer ( 20 mM HEPES , pH 7 . 4 , 150 mM NaCl ) at a final lipid concentration of 2 mM and incubated for 10 min at room temperature with peptides ( 4 µM ) . Liposome-protein complexes were recovered by centrifugation ( 100 , 000×g ) at room temperature for 20 min , the supernatant was completely removed , and sedimented liposomes were solubilized in SDS sample buffer . The peptides in the supernatant and pellet were subjected to SDS-PAGE using 16% tricine gels , and analyzed as described above . Jurkat T cells expressing GFP fusion proteins were fixed in 0 . 5% glutaraldehyde and 4% paraformaldehyde in 0 . 05 M sodium cacodylate buffer ( pH 7 . 2 ) at 4°C for 2 h . Ultrathin sections ( 50 nm in thickness ) were cut using an ultramicrotome ( MT-X; RMC ) and stained with an anti-GFP primary antibody and an anti-rabbit IgG secondary antibody conjugated with 10 nm gold particles ( Sigma ) . Sections were then stained with 2% uranyl acetate and Reynolds' lead citrate , and examined by transmission electron microscopy ( LIBRA 120; Carl Zeiss ) at an accelerating voltage of 120 kV . Negative control experiments were also performed to ensure the specificity of the labeling by replacing the primary antibody with rabbit preimmune serum . Liposomes incubated with peptides as described in liposome binding assays were adsorbed onto carbon-coated copper grids , stained with uranyl acetate , and then observed by electron microscopy . The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) accession numbers for Tip and Tio proteins used in this paper are Tip C488 ( AAA72928 ) , Tip C484 ( P88825 ) , Tip C484-77 ( P25049 ) , and Tio ( AAC95538 ) .
|
Herpesvirus persists in its host by entering a latent state , periodically reactivating to produce infectious viral particles . Some of the herpesviruses have also been known to be related to cancers . Herpesvirus saimiri ( HVS ) , an oncogenic monkey herpesvirus , persists in the T lymphocytes of its natural host , the squirrel monkey , without any apparent disease symptoms , but infection of other species of New World and Old World primates results in fulminant T cell lymphomas . Two viral oncoproteins , Saimiri Transforming Protein and Tyrosine kinase-interacting protein ( Tip ) , are required for T cell transformation . It has been known that Tip may also play some role in viral persistency within T cells by inhibiting the activation of the host cells upon antigenic stimulation . Here , we have identified a structural domain , a putative amphipathic helical motif , preceding the transmembrane domain of Tip . We also found that the structural motif is essential for Tip's localization on specialized membrane domains , lipid rafts , and selective downregulation of antigen receptors . Furthermore , we could genetically dissect the functional roles of the amphipathic helical motif and transmembrane domain of Tip in membrane deformation and oligomerization , respectively . These findings significantly advanced our understanding of how herpesvirus modulates host lymphocytes for viral persistence and pathogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/persistence",
"and",
"latency",
"cell",
"biology/membranes",
"and",
"sorting",
"biochemistry/cell",
"signaling",
"and",
"trafficking",
"structures",
"virology/viruses",
"and",
"cancer",
"virology/immune",
"evasion"
] |
2008
|
Role of Amphipathic Helix of a Herpesviral Protein in Membrane Deformation and T Cell Receptor Downregulation
|
Cholera is an acute diarrheal disease and a major public health problem in many developing countries in Asia , Africa , and Latin America . Since the Bay of Bengal is considered the epicenter for the seventh cholera pandemic , it is important to understand the genetic dynamism of Vibrio cholerae from Kolkata , as a representative of the Bengal region . We analyzed whole genome sequence data of V . cholerae O1 isolated from cholera patients in Kolkata , India , from 2007 to 2014 and identified the heterogeneous genomic region in these strains . In addition , we carried out a phylogenetic analysis based on the whole genome single nucleotide polymorphisms to determine the genetic lineage of strains in Kolkata . This analysis revealed the heterogeneity of the Vibrio seventh pandemic island ( VSP ) -II in Kolkata strains . The ctxB genotype was also heterogeneous and was highly related to VSP-II types . In addition , phylogenetic analysis revealed the shifts in predominant strains in Kolkata . Two distinct lineages , 1 and 2 , were found between 2007 and 2010 . However , the proportion changed markedly in 2010 and lineage 2 strains were predominant thereafter . Lineage 2 can be divided into four sublineages , I , II , III and IV . The results of this study indicate that lineages 1 and 2-I were concurrently prevalent between 2007 and 2009 , and lineage 2-III observed in 2010 , followed by the predominance of lineage 2-IV in 2011 and continued until 2014 . Our findings demonstrate that the epidemic of cholera in Kolkata was caused by several distinct strains that have been constantly changing within the genetic lineages of V . cholerae O1 in recent years .
Cholera is an acute life-threatening diarrheal disease and remains a major health threat , particularly in developing countries in Asia , Africa , and Latin America [1 , 2] . It is estimated that 1 . 4 to 4 . 3 million cases of cholera and 28 , 000 to 142 , 000 deaths due to cholera , occur every year worldwide [3] . The Gram-negative bacterium Vibrio cholerae has more than 200 serogroups , but only O1 and O139 serogroups are responsible for epidemic and pandemic cholera [4 , 5] . Serogroup O1 has been further classified into two biotypes , classical and El Tor , based on several phenotypic traits . Each biotype has a unique nucleotide sequence within the genes encoding the cholera toxin responsible for severe diarrhea [4] . The cholera toxin is encoded on the lysogenic bacteriophage CTXΦ and consists of two subunits , A and B [6] . Corresponding amino acids at 20/39/68 of the cholera toxin B subunit of the classical biotype are H/H/T ( ctxB1 genotype ) , and of the El Tor biotype are H/Y/I ( ctxB3 genotype ) [7] . Historically , seven cholera pandemics have been recorded since 1817 . The sixth , and presumably earlier pandemics , emerged from the Bay of Bengal and were caused by the V . cholerae O1 classical biotype . However , the current seventh pandemic is caused by the El Tor biotype [5] . Recently , Hu et al . used comparative genomic analysis to demonstrate that the seventh pandemic strains originated from a nonpathogenic strain first observed in 1897 and slowly acquired virulence-associated elements by 1954 before becoming pandemic in 1961 [8] . Over the years , the V . cholerae O1 El Tor biotype has shown remarkable change and developed novel pathogenic variants that have the classical type ctxB gene ( ctxB1 ) with an El Tor type genomic backbone [9–15] . Recently , the new ctxB variant ( ctxB7 ) was found in Haiti , amongst other countries , which has N/H/T at the amino acid position 20/39/68 of the cholera toxin B subunit [16] . This newly appeared variant of V . cholerae has totally replaced the old strains indicating that the predominant strain has shifted during the current pandemic [16] . It has also been reported that the Haitian variant strain has evolved due to sequential events in the Indian subcontinent with some cryptic modification in the genome [17] . Mutreja et al . reported that the seventh pandemic strain first appeared in the Bay of Bengal and recurrently spread from this area to different parts of the world in at least three waves [18] . The Bay of Bengal is therefore considered the epicenter for the seventh cholera pandemic . It is important to monitor the appearance of new variants of V . cholerae in the Bay of Bengal , as it is possible that they will spread around the world in the future . From the active diarrheal disease surveillance in the Infectious Diseases and Beliaghata General Hospital in Kolkata , it was established that V . cholerae O1 is one of the most common bacterial pathogens associated with diarrhea , with an estimated 11 , 000 cases every year [19] . Thus , cholera continues to be an important public health problem; hence , it is important to understand the genetic dynamism of V . cholerae from Kolkata , as a representative area of the Bengal region . Vibrio Seventh Pandemic Island ( VSP ) was first detected by comparative genomic analysis of the classical and El Tor biotype strains of V . cholerae O1 [20] . Although the two genomic regions , VSP-I and VSP-II , were identified to be unique in the seventh pandemic El Tor strains , the role of these genomic islands in the pathogenicity of the organism is yet to be established . VSP-II is a 26 . 9-kbp genomic region composed of 24 genes between VC0490 and VC0516 according to the annotation of whole genome sequence ( WGS ) of V . cholerae N16961 . These include genes encoding RNase , type IV pilin , chemotaxis , DNA repair , and transcriptional regulator [21] . Several variants of VSP-II have been reported in El Tor strains isolated from different continents , including Asia , Africa , and Latin America [22–26] . Therefore , characterization of VSP-II types is helpful in understanding the genetic lineages involved in the global transmission of cholera . In addition , WGS analysis is currently used as a powerful tool for understanding the various functional and evolutionary aspects of the organisms [27–32] . In this study , we carried out WGS analysis of V . cholerae O1 strains isolated between 2007 and 2014 from cholera patients in Kolkata in order to determine their genetic lineages . This analysis revealed the heterogeneity of VSP-II in Kolkata . In addition to the VSP-II genotype , phylogenetic analysis based on the whole genome single nucleotide polymorphisms ( SNPs ) revealed that shifts of predominant strains have occurred several times in recent years in Kolkata .
To understand the genomic diversity of V . cholerae O1 , we randomly selected 10 strains isolated from hospitalized cholera patients in each year between 2007 and 2014 . Draft genome sequences of 80 strains were obtained using a next generation sequencer . Sufficient DNA sequencing reads were generated to cover the genome at least 60 folds in 79 strains . One strain was excluded from further analysis due to poor quality of the sequence data . The short sequence reads were mapped onto the genomic sequence of V . cholerae N16961 as an El Tor reference strain . We found that all strains lacked part of VSP-II , which is a 26 . 9-kbp genomic island consisting of 24 open reading frames ( ORFs ) , VC0490 to VC0516 ( Fig 1A ) . We determined the genetic organizations of VSP-II in these strains . The region from 119th nucleotide of VC0495 to 1320 bp downstream of the VC0498 stop codon was replaced by a 1257 bp DNA fragment consisting of genes for transposase A and B subunits in 19 isolates ( VSP-IIB , Fig 1B ) . Furthermore , 2 out of these 19 strains had an insertion of a 1767-bp DNA fragment of SXT IS4 family transposase gene at the 510th nucleotide of the VC0516 gene ( VSP-IIBv1 , S1 Fig ) or at the 780th nucleotide of the VC0515 gene ( VSP-IIBv2 , S1 Fig ) . Fifty-eight strains had a larger deletion than VSP-IIB . The 119th nucleotide of VC0495 to 596 bp downstream of the VC0512 stop codon ( 14 , 376 bp length ) was replaced by genes for transposase A and B subunits ( VSP-IIC , Fig 1C ) . Moreover , 9 of these strains had an insertion of the additional transposase gene fragment at the 560th nucleotide of VC0492 ( VSP-IICv1 , S1 Fig ) . The deleted regions are distinct in VSP-IIB and VSP-IIC; however , the upstream terminals of the replaced area ( 119th nucleotide of VC0495 ) were identical . These observations may suggest the existence of a hot spot for transposase . Two strains were negative for VSP-II ( VSP-IID , Fig 1D ) . The VSP-II was integrated between two attachment sites , attL ( 14 bp ) and attR ( 16 bp ) , in all the VSP-II positive strains ( Fig 1A ) . Murphy et al . demonstrated that VSP-II could be excised from the chromosome due to VC0516 , which encodes an integrase , and the post-excision att site was of a shorter type ( 14 bp ) , identical to attL [32] . Classical strains O395 and V51 , which are VSP-II-negative isolates , have a 16-bp att site identical to attR . However , VSP-II-negative strains in this study had an att site of 13 bp ( Fig 1D ) . A shortened att site might be a consequence of excision of VSP-II from a VSP-II-positive El Tor strain . Taken together , we identified three types of VSP-II in clinical V . cholerae O1 isolates in Kolkata ( Fig 1 ) . In addition , variants of VSP-IIC that have insertions of an additional transposase gene fragment in VC0492 , and of VSP-IIB , that also has an insertion of SXT IS4 family transposase gene in VC0516 or VC0515 , were identified ( S1 Fig ) . Frequencies of three VSP-II types in each year are shown in S2 Fig . Interestingly , VSP-IIC strains numbered less than half until 2009 . This genotype rapidly spread in 2010 and eventually replaced the other types during subsequent years ( S2 Fig ) . These results indicate that the genomic islands of V . cholerae O1 strains are frequently rearranged in Kolkata . VSP-IICv1 strains have an additional transposase gene fragment inserted into VC0492 ( S1 Fig ) . Nine strains were identified as this variant , all of which were isolated in 2011 , indicating that VSP-IICv1 strains appeared in 2011 and suddenly became predominant , but then quickly disappeared in 2012 . The results also indicate that the predominant strain of V . cholerae O1 can rapidly shift in a particular area during an endemic , or that diverged strains were present in the environment of a particular area , which might have caused the seventh pandemic . V . cholerae O1 El Tor strains were characterized by several ctxB variants including ctxB3 , ctxB1 , and ctxB7 in the typical El Tor type , El Tor variant , and Haitian variant , respectively . Genotypes of ctxB and VSP-II of each strain used in this study are shown in S1 Table . Although strains were isolated from hospitalized patients with typical cholera symptoms , two strains were negative for ctxB ( strain IDH-00115 in 2007 and IDH-02185 in 2009 ) . Our hospital-based surveillance screened for 25 enteric pathogens including 15 bacteria , 6 viruses and 4 parasites in each fecal sample [19] . However , the ctxB negative V . cholerae O1 was the sole pathogen detected . Although epidemic cholera is caused by cholera toxin-positive V . cholerae , strains without cholera toxin can cause a diarrheal disease through other possible virulence factors , including the heat-stable toxin ( NAG-ST ) [33] , hemolysin ( Hly ) , type III secretion system ( T3SS ) [34 , 35] , cholix toxin ( Chx ) [36 , 37] , mannose sensitive hemagglutination ( MshA ) and repeat in toxin ( RtxA ) . Two ctxB-negative strains , as well as the ctxB-positive strain in current study , harbored genes encoding for Hly , MshA and RxtA , but not NAG-ST , structural proteins of T3SS and Chx . The other 77 strains harbored the ctxB gene , either ctxB1 ( n = 20 ) or ctxB7 ( n = 57 ) genotypes . As shown in S1 Table , 18 strains with ctxB1 had VSP-IIB and the other 2 strains had VSP-IID . All 57 strains with ctxB7 had VSP-IIC . Both VSP-II and CTXΦ prophage are mobile elements and the distance between these elements is more than 1 Mbp on the 2 . 96 Mbp on chromosome 1 in the reference genome . If the mutations in ctxB and VSP-II were independent events , either element may have affected the acquisition or selectivity of another element . To assess the genetic lineage of V . cholerae O1 clinical isolates in Kolkata , we performed phylogenetic analysis based on the genome-wide SNPs . As shown in Fig 2A , the seventh pandemic El Tor strains clade differed from the pre-seventh pandemic strains . The 79 V . cholerae strains isolated in Kolkata between 2007 and 2014 belonged to wave 3 of seventh pandemic and were classified into two lineages ( Fig 2B , blue and red branches ) . Lineage 1 contained 20 Kolkata strains with the ctxB1 allele , one Kolkata strain negative for ctxB , and other strains isolated in India and Nepal . In addition to 19 VSP-IIB strains , the two VSP-IID strains were also found in this lineage . This result is consistent with the notion that these VSP-IID strains are not of the classical biotype but instead seventh pandemic strains after the excision of VSP-II , as suggested by the short att site . Each VSP-IID strain is closely related to each other and with the VSP-IIB strains , suggesting that VSP-IID is a derivative of the VSP-IIB strain as a consequence of the excision of VSP-II . Lineage 2 contains 58 strains , of which 57 strains possesses ctxB7 and one negative for ctxB gene . Although all 58 lineage 2 strains harbored VSP-IIC , lineage 2 was divided into four sublineages , I , II , III , and IV ( Fig 3 ) . Lineages 2-I , 2-II , and 2-III comprised South Asian isolates; however , 2-IV also contained Haitian isolates . In this lineage , Kolkata strains isolated between 2010 and 2014 were more clustered among themselves owing to the relatedness between Nepalese and Haitian isolates ( Fig 3 ) . Strains with VSP-IICv1 , which is a transposon-inserted variant of VSP-IIC , formed a cluster ( Fig 3 ) , suggesting clonal expansion of the lineage 2-IV with VSP-IICv1 . Each of the two ctxB-negative strains was phylogenetically within lineages 1 and 2 , therefore seeming to emerge independently . Phylodynamic analysis of the 79 Kolkata strains was investigated via Bayesian analysis ( Fig 4 ) . Kolkata strains between 2007 and 2014 were divided into lineage 1 ( n = 21 ) , which possesses VSP-IIB and ctxB1 , and lineage 2 ( n = 58 ) that largely possesses VSP-IIC and ctxB7 . It was estimated from the maximum clade credibility tree that the most recent common ancestor ( MRCA ) of lineage 2 emerged in July 2006 ( 95% HPD: September 2005 to February 2007 ) . Lineage 1 emerged in January 2006 ( 95% HPD: January 2005 to October 2006 ) and the isolates in lineage 1 lasted until December 2010 . Consequently , two distinct V . cholerae O1 lineages were concurrently distributed in Kolkata between July 2006 and December 2010 . As for the 58 strains in lineage 2 , Kolkata strains belonged to lineage 2-I ( n = 9 ) , 2-III ( n = 8 ) , or 2-IV ( n = 41 ) . The isolates of lineage 2-I that circulated between 2007 and 2009 were replaced with the isolates of lineage 2-III . The analysis predicted that the MRCA of lineage 2-III existed in August 2009 ( 95% HPD: April 2009 to December 2009 ) ; however , lineage 2-III was transient , at least in Kolkata , because the isolates were only observed in 2010 . Lineage 2-IV strains followed the transient lineage 2-III spike in Kolkata , for which the divergence time of the two lineages was estimated to be in March 2008 ( 95% HPD: July 2007 to November 2008 ) . All 40 strains between 2011 and 2014 belonged to the lineage 2-IV cluster with Nepalese and Haitian isolates , and the MRCA seemed to emerge in April 2010 ( 95% HPD: September 2009 to September 2010 ) . The first lineage 2-IV strain isolated in March 2010 was different from the other lineage 2-IV Kolkata strains and assigned to a distinct cluster of Indian isolates , which were isolated in Northern India in 2009 [23] , suggesting that a surge in lineage 2-IV including Haitian isolates in Kolkata started suddenly at April 2010 and continued until 2014 .
WGS of clinical V . cholerae O1 isolates in Kolkata determined the sequence variation of VSP-II , which related to the ctxB allele . The phylogenetic analysis of strains found in Kolkata revealed two distinct lineages ( lineages 1 and 2 ) and the coexistence of strains of both lineages between July 2006 and December 2010 , thus indicating the concurrent prevalence of at least two genetically distinct V . cholerae strains . However , the ratio of the two lineages changed markedly from 2010 onward . Lineage 2 strains increased in 2010 and totally replaced lineage 1 strains in 2011 , which continued to be predominant until 2014 . Additionally , strains in lineage 2 were diverse and showed a temporal pattern . Lineage 2 strains isolated between 2007 and 2009 were categorized as lineage 2-I and those observed during 2010 , lineage 2-III . Strains in lineage 2-IV were first found in 2010 and then became predominant . Around the same time elsewhere in India , V . cholerae strains showed variations in several genes and seemed to evolve sequentially with some cryptic modification in the genome [17 , 38 , 39] . Our results are in agreement with previous findings of genome-wide SNP analysis , suggesting that the genotypes of V . cholerae O1 in Kolkata had been replaced on several occasions in recent years . Since the discovery of the prototypical VSP-II genomic island in 2004 [21] , several variants have been identified from different continents . An environmental isolate in Brazil in 1982 , TMA21 , had deletions from downstream of VC0498 to VC0503 and from VC0511 to VC0515 [26] . Clinical isolates in Peru between 1991 and 2003 lacked genes VC0512 to VC0515 [25] , and West African and South American isolates between 1981 and 1985 also lacked VC0512 to VC0515 [18] . In Africa , Zambian isolates from 2003 to 2004 had deletions from VC0493 to VC0498 [23] . However , such VSP-II variants were not found in Kolkata between 2007 and 2014 . CIRS101 isolated in Bangladesh in 2002 had a substitution from VC0495 to VC0512 by transposase , which is identical to VSP-IIC in this study [26] . Moreover , both VSP-IIB isolates and VSP-IIC isolates were found in Chandigarh , a province of northern India , in 2009 [22] . These strains could emerge and spread widely throughout the Indian subcontinent and , further work with retrospective analysis would be required to elucidate the emergence mechanism of VSP-II variants . It has been reported that V . cholerae strains in all pandemics disseminated from the Bay of Bengal to the rest of the world [18] and considering this tendency , it is possible that the new VSP-II variants could spread beyond this region . Taviani et al . reported that among 97 isolates in Bangladesh between 2004 and 2007 , 96 strains harbored VSP-IIC [26] . This type was found to be predominant in Kolkata after 2010 ( S2 Fig ) . Although both Kolkata and Bangladesh border the Bay of Bengal , transition patterns of predominant strains are temporally distinct . In addition , prevalence of the ctxB allele also differed between Dhaka , Bangladesh , and Kolkata . In Dhaka , isolates with ctxB1 reemerged in 2012 and became dominant between 2013 and 2014 by outcompeting the former dominant ctxB7 strains [40] . During the same period in Kolkata , all strains possessed the ctxB7 allele [17] and belonged to lineage 2-IV in this study . From our SNP analysis and correlation with ctxB typing , we speculate that the current predominant strain in Dhaka belongs to lineage 1 . Considering the prevalence trends of V . cholerae O1 in Kolkata , novel genetic variants may appear frequently and spread to other regions . Our genome-wide SNP analysis demonstrates the phylogenetic relatedness between Kolkata strains and strains isolated from other areas , especially strains in lineage 1 . However , the lineage 2 strains formed a spatiotemporal homogeneous cluster . The limited amount of available genome data might affect the apparent homogeneous cluster formation . Two exceptions are observed in clusters in lineage 2-III , formed by Kolkata and Nepalase strains , and in lineage 2-IV . The former , a cluster made by 8 Kolkata strains isolated in 2010 includes one Nepalese 2010 strain . The other cluster , consisting of Northern India strains isolated in 2009 , includes one Kolkata strain isolated in 2010 . More genomic data from next-generation sequencing would reveal more precise dissemination and evolutionary trends of V . cholerae O1 . WGS-based analysis could help us to understand the temporal and geographical spread of V . cholerae; hence , continued monitoring of V . cholerae O1 is needed in all cholera endemic regions . In addition , WGS has been utilized in several studies to understand global transmission and phylogeny of pathogenic bacteria including V . cholerae , Shigella dysenteriae and Salmonella Enteritidis [18 , 41 , 42] . Our work characterized the transition of predominant strains during several continuous years at the epicenter of cholera . Combining these studies with computational modeling may enable us to predict strains that cause epidemics throughout the world .
Vibrio cholerae O1 strains used in this study are listed in S1 Table with the year and month of isolation . These strains were isolated from fecal samples of hospitalized patients with typical cholera symptoms in Kolkata , India between 2007 and 2014 . V . cholerae strains were isolated by streaking the stool samples on thiosulphate citrate bile salts sucrose ( TCBS ) agar plates and typical sucrose fermenting yellow colonies were tested by serum agglutination using V . cholerae O1 polyvalent antiserum ( Becton Dickinson , Sparks ) . Isolated strains were stored in -80°C as glycerol stock . Fecal samples were collected two days a week from every fifth diarrheal patient ( approximately 5 . 7% of total diarrheal patients ) . Annually , 93 to 363 samples were positive for V . cholerae O1 between 2007 and 2014 . Ten V . cholerae O1 strains were chosen each year at random to represent predominant months of each year and subjected to the analysis . The patients were aged between 1 month and 89 years at two hospitals in Kolkata , and all patients were discharged after treatment . Genomic DNA was prepared using the DNeasy Blood & Tissue Kit ( Qiagen ) according to the manufacturer’s instructions . In total , 80 V . cholerae O1 strains were used for whole genome sequencing . We prepared Illumina libraries using Nextera XT DNA Library Preparation Kit ( Illumina ) and sequenced paired-end Illumina short reads for each library on HiSeq 2500 ( Illumina ) or MiSeq ( Illumina ) sequencers . The sequence reads were mapped with V . cholerae O1 El Tor reference strain N16961 by using CLC Genomics Workbench 8 . 5 . 1 ( CLC bio ) to obtain whole genome alignment . The reads were assembled using a de novo genome assembly program of CLC Genomics Workbench and a multicontig draft genome was generated for each sample . Nucleotide sequence data were submitted to the DDBJ Sequenced Read Archive and each accession number is listed in S1 Table . Comparison of short sequence reads with the genomic sequence of V . cholerae N16961 , a prototype strain of seventh pandemic El Tor biotype , suggested the lack of a large area in the VSP-II genomic island; thus , genetic organizations of VSP-II in each strain used in this study were identified ( Fig 1 ) . Among 79 strains , 19 , 58 , and 2 were found to lack the ORFs from VC0495 to VC0498 , VC0495 to VC0512 , and the entire VSP-II sequence , respectively . Strains lacking the entire VSP-II sequence generated a contig cover from upstream to downstream of VSP-II region ( Fig 1D ) . Nineteen strains lacked the internal region of VSP-II from VC0495 to VC0498 as compared with the N16961 sequence and did not generate a contig cover from upstream to downstream of the missing area . This area of each strain was PCR amplified and the sequence of this amplicon was determined using the Sanger method ( Fig 1B ) . Among these 19 strains , 2 strains had truncated contigs within VSP-II region in addition to the lacking area . These contigs were also amplified and the sequence was determined ( S1 Fig ) . The rest of the 58 strains were found to lack the ORFs from VC0495 to VC0512 as compared with the N16961 sequence . This area of each strain was PCR amplified and determined the sequence of a representative strain ( Fig 1C ) . To confirm the identity of VSP-II , the entire VSP-II fragments of these strains were amplified and analyzed by Restriction Fragment Length Polymorphism ( RFLP ) using BglII . Nine strains showed altered band patterns compared with the others , and the difference of these VSP-II sequence were identified by PCR amplification and sequencing ( S1 Fig ) . To remove adapter sequences and low quality bases with a Phred score of less than 15 from the short reads , read trimming was performed using fastq-mcf ( https://expressionanalysis . github . io/ea-utils/ ) and sickle ( https://github . com/najoshi/sickle ) program . Simulated paired-end reads were constructed from the available genomic sequences of V . cholerae strains using SimSeq software [43] with the following parameter: number of pairs of reads , ‘‘read_number 2000000”; mean library insert size , ‘‘insert_size 150”; and paired-end reads length of 120 mer , ‘‘21 120 22 120” . These parameters indicated that 4 million hypothetical 120-mer reads were generated without mutations or indels from the genomic sequences used for SNP identification . The trimmed or simulated reads with at least 40 mer were mapped using the BWA-mem program [44] against the N16961 sequence ( NC_002505 . 1 and NC_002506 . 1 ) [45] , and the read mapping data were constructed by the samtools program [46] . All SNPs with at least a 5× coverage depth and a Phred score of at least 20 were extracted using VarScan v2 . 3 . 4 [47] . The SNPs on the repeat and prophage regions of the N16961 genome , which was identified by the NUCmer [48] and PHAST [49] program , was excluded for further core-genome phylogeny analysis . Additionally , to exclude for the recombination regions , RecHMM was used to identify the recombination region [50] . The remaining SNPs were concatenated to generate a pseudo sequence for phylogenetic analysis; maximum likelihood phylogenetic analysis was performed using RAxML v8 . 2 . 0 [51] with 1 , 000 bootstrap iterations . The trees were visualized using iTOL 3 ( http://itol . embl . de/ ) [52] . To estimate the divergence date of the V . cholerae O1 isolate in Kolkata , we performed a temporal analysis using the Bayesian Evolutionary Analysis Sampling Trees ( BEAST ) v . 2 . 4 . 4 software package [53] . The isolation date of each strain was used as tip data . A random clock model was implemented using Markov Chain Monte Carlo ( MCMC ) chains run for 100 million generations with 10% burn-in and sampled every 1000 generations . A GTR nucleotide substitution model was used with a gamma distribution with four rate categories . The effective sample sizes were greater than 200 for all estimated parameters and tree data were summarized to generate the maximum clade credibility tree . This study was approved by the duly constituted Institutional Ethics Committee ( IEC ) of National Institute of Cholera and Enteric Diseases . As per the recommendation of IEC , individual written informed consent was obtained from each adult patient and a parent or guardian of child patient enrolled in this study and confidentiality was maintained .
|
Seven cholera pandemics have been recorded throughout history , and the sixth , and presumably earlier pandemics , emerged from the Bay of Bengal . The seventh pandemic strain also appeared and spread from this area to different area of the world . Thus , the Bay of Bengal has always been considered the epicenter of cholera pandemics . In this report , we characterized the V . cholerae strains isolated from patients with cholera in Kolkata as a representative area of the Bay of Bengal between 2007 and 2014 . The analysis revealed that the cholera epidemics were caused by several distinct V . cholerae O1 strains and that the predominant strains have genetically changed several times in recent years .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"vibrio",
"tropical",
"diseases",
"microbiology",
"toxicology",
"toxic",
"agents",
"bacterial",
"diseases",
"phylogenetics",
"vibrio",
"cholerae",
"data",
"management",
"phylogenetic",
"analysis",
"genome",
"analysis",
"neglected",
"tropical",
"diseases",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"cholera",
"genomics",
"bioinformatics",
"medical",
"microbiology",
"microbial",
"pathogens",
"comparative",
"genomics",
"evolutionary",
"systematics",
"database",
"and",
"informatics",
"methods",
"el",
"tor",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"organisms"
] |
2017
|
Comparative genome analysis of VSP-II and SNPs reveals heterogenic variation in contemporary strains of Vibrio cholerae O1 isolated from cholera patients in Kolkata, India
|
Dengue in Venezuela is a major public health problem with an increasing incidence of severe cases . Early diagnosis and timely treatment influences the outcome of dengue illness , as delay in care-seeking is significantly associated with complications leading to severe dengue . We aimed to understand patterns of health seeking behaviour ( HSB ) in individuals exposed to high dengue incidence in order to improve early attendance to health centres . Between September 2013 and February 2014 a cross-sectional household survey was performed in Maracay , Venezuela . Intended HSB of adults and children’s parents/guardians was assessed with respect to fever or suspected dengue . Data was collected through structured questionnaires from 105 individuals . Most individuals felt at risk of dengue and believed it could be a deadly disease . In the case of suspected dengue , the majority ( 60% ) would choose to first seek medical help versus first treating at home , in contrast to 11% in the case of fever . Amongst those who decided to visit a doctor , a suspected dengue infection would prompt them to search medical help earlier than if having only fever ( p<0 . 001 ) . Multivariate analysis modelling showed that the independent factors associated with the intention to firstly visit a doctor versus treating at home in the case of dengue were feeling at risk ( OR = 3 . 29; p = 0 . 042 ) and being an adult ( as opposed to caring for a child as a parent/guardian; OR = 3 . 33 , p = 0 . 021 ) , while having had a previous dengue infection ( OR = 0 . 29; p = 0 . 031 ) and living in the neighbourhood Caña de Azúcar ( OR = 0 . 28 , p = 0 . 038 ) were negatively associated with seeking medical care as their first action . Knowledge of HSB related to dengue is scarce in the Americas , our study attempts to contribute to a better understanding of HSB in this region . Improving early dengue disease recognition and awareness may enhance prompt attendance to medical care in affected populations and thereby reduce mortality and severity of dengue . Especially for those with a previous dengue infection , efforts have to be made to promote prompt health centre attendance .
Dengue fever , a viral vector-borne disease spread by the day-biting mosquitoes Aedes aegypti and A . albopictus , is a global health problem of increasing importance [1] . Currently , dengue affects over 2 . 5 billion people living in dengue endemic areas , which comprises 40% of the world’s population [2] . According to estimations of the WHO , 50–100 million dengue infections occur every year , leading to 500 000 cases of severe disease that need hospitalisation [2] . However , recent estimations speak of approximately 400 million dengue infections annually [3] . Where in the 1950s dengue cases were reported in only nine countries , today more than 125 countries in the tropics and subtropics are endemic for dengue [4] . In the Americas , almost all countries struggle with recurrent epidemics [5] . The poverty , poor sanitation and overcrowding that accompanies the uncontrolled urbanisation in this region creates environments in favour of vector-breeding and rapid spread of the virus , which leads to serious obstacles in disease control [6] . Dengue has become a major public health problem in Venezuela , with epidemics of increasing magnitude regularly occurring against a background of an established endemic situation . Initial descriptions of dengue-like illness in Venezuela based on clinical manifestations date from 1828 and 1946 [7] . Since the first dengue hemorrhagic fever epidemic reported in the country in 1989–1990 and the second in the Americas [8] , the frequency of severe cases has increased . Between 1989 and 2007 , the highest proportion ( 35% ) of severe dengue cases within the Americas were reported in Venezuela [9] . In effect , dengue transmission in Venezuela has become perennial with poverty-related socio-economic factors and behavioural determinants fuelling the increasing incidence of dengue in the urban areas of the country [9 , 10] . The most recent and largest dengue outbreak took place in 2010 with more than 120 . 000 reported cases , of which 8% represented severe cases [11] . Early diagnosis and adequate supportive care are of great importance in the management of dengue so as to avoid the development of complications and severe disease . Thereby , early treatment intervention can reduce the case fatality rate from 20% to 1% or less [1 , 12] . While knowledge and possibilities to diagnose and treat dengue fever increase , efforts have to be made to make these new developments accessible for those who have a dengue infection . An important factor to be taken into consideration is the patient’s health seeking behaviour ( HSB ) , because for early diagnosis and supportive care , people must have the intention and the means to seek medical care early in the disease . Therefore , local studies on health believes and practices , HSB and access to care with respect to dengue fever are needed to identify barriers and opportunities for applying these new developments in diagnostics and treatment [13] . Insights in HSB of dengue could help to attain a reduction of late diagnosis , an increase of treatment adherence and improvement of health promotion strategies applied to a specific culture [14] . Two important health behaviour theories are the Health Believe Model ( HBM ) [15] and the Theory of Planned Behaviour ( ToPB ) [16] . A central concept in the HBM is the ‘perceived susceptibility’ , which refers to the perceived chance of acquiring a condition ( in this article also referred to as ‘risk perception’ ) . The ‘perceived susceptibility’ and the ‘perceived severity’ leads to the formation of ‘perceived threat’ of a certain condition . The likelihood of performing a certain health behaviour is directly linked to a ) the perceived threat , b ) the perceived benefits and barriers of the suggested behaviour change , c ) the self-efficacy and d ) the cues to action [15 , 17] . The ToPB links the attitude towards behaviour , subjective norms and perceived behavioural control to behavioural intentions and behaviour [16 , 18] . In Venezuela , patients with a suspected dengue infection tend to seek medical help beyond the third day after the onset of fever [19] . At this time , the patient may be already critically ill [20] . Delay in care-seeking is found to be significantly associated with complications leading to severe dengue [21] , which stresses the importance of understanding HSB and access to care for dengue patients . However , research on HSB applied to dengue appears to be scarce , especially in the Americas , as the majority of studies addressing ( partially ) this topic have been performed in Asia [22–28] . Moreover , the published literature on HSB related to dengue was not [22–25 , 27 , 28] , or was only partly [26] based on health behaviour theories . This study aims to understand the patterns of HSB in the Venezuelan population exposed to high dengue incidence in order to find ways to improve early attendance to health centres and medical care . We compared HSB intentions of adults and of parent/guardians with respect to their children in the case of fever or suspected dengue . By using quantitative data supplemented with qualitative data based on the HBM , we aim to present a better insight in social , psychological and cultural motives of the intended behaviour and attitudes .
In August 2010 a prospective , community-based cohort study was set up in Maracay , Venezuela . Maracay is one of the largest cities of Venezuela with dengue hyper-endemicity [9 , 29] . It is the capital of Aragua state with an estimated of 1 . 300 . 000 inhabitants [30] . Aragua state witnessed the highest incidence of dengue in Venezuela in 2012 , reaching nearly 7000 reported cases of which 2% were severe [11] . The study site and design has been described elsewhere [10] . Briefly , three neighbourhoods within Maracay called Candelaria , Caña de Azúcar and Cooperativa were selected for their high dengue incidence . All neighbourhoods are served by public ( governmental primary and secondary ) health centres [11] . . Patients that require further specialised treatment are referred to the main public tertiary level hospital , the Hospital Central de Maracay . Private hospitals and clinics are also part of the health system . Within the cohort study , 2014 individuals aged 5–30 years old living in 840 households were enrolled at the start of the study and followed annually [12] . The present study was performed during the September 2013 and February 2014 annual survey . A cross-sectional survey of a sub-sample of the cohort participants was carried out to gather quantitative and qualitative data on HSB intentions at community level of the general population exposed to dengue . The study was conceptualised to first address the interviewees’ HSB with respect to fever only as a symptom of several possible diseases . Subsequently , individuals were enquired about their HSB in regard to a specific disease , in this case a suspected dengue infection . Therefore , the study uses aspects of the HBM theory to incorporate the concepts of susceptibility and severity in the analysis to understand the HSB after the onset of fever as this symptom could lead to the individual perceiving susceptibility to multiple conditions [15 , 17] . This theory was not applied in the data collection but aided in the analysis and interpretation of the results . A randomized sub-sample of approximately 260 households included in the cohort study was selected . One individual was interviewed in each household . The intention was to interview an equal number of adults and parents or guardians of children ( <18 years old ) who were already participating in the cohort study . Adults ( 18 years and older ) were randomly chosen from all present adults at the moment of visiting the selected households . A structured questionnaire , the HSB-questionnaire , was developed containing pre-coded and open questions on socio-demographic and socio-economic details , knowledge of dengue symptoms and dengue transmission , risk perception , pathways of HSB in relation to presenting fever and suspicion of dengue infection . With respect to the parents/guardians interviewed , questions on HSB and risk perception referred to the child , while adults were interviewed with respect to their own attitudes and practices . The questionnaires were prepared in English , translated to Spanish , pre-tested and adapted in a pilot study . Data on socio-economic variables were collected from a household questionnaire which was applied as part of the annual survey of the cohort study . Information collected in the questionnaires was entered into a database using Epi Info ( Epi Info , version 3 . 5 . 4 ) . Data was checked for consistency and analysed anonymously . Chi-square test or Fisher’s exact test were used to assess proportions . Continuous variables were converted into ordered categorical variables when suitable . For normally distributed quantitative data , means were compared using Student’s t-test otherwise , the Mann-Whitney U test was used . The Wilcoxon signed rank test was used for comparing related means within individuals when comparing HSB of fever and dengue while pair-wise proportions were compared with a McNemar’s test . Significance was determined at 5% level . Principal components analysis [34 , 35] was utilised to weigh socio-economic variables , obtain a relative score and classify individuals into low , average and high socio-economic status . Logistic regression was used to compare crude and adjusted odds ratios ( OR ) . Multivariate logistic regression analysis was used to determine variables independently associated with the intended first action reported in HSB pathways for fever and suspected dengue . Variables with a p-value ≤0 . 2 after adjusting by age group and sex were fitted into these multivariate models and adjusted for further confounders . Effect modification was analysed and resulting models compared by a likelihood ratio test . Two separate final models are presented , one in the case of fever and the second in the case of intended suspected dengue ( Table 1 ) . Data was analysed using SPSS ( SPSS Inc . , version 20 . 0 , Chicago , Illinois ) and STATA ( Stata Statistical Software: Release 10 . College Station , TX , ) softwares . This study was approved by the Ethics Review Committee of the Biomedical Research Institute , Carabobo University ( Aval Bioetico #CBIIB ( UC ) -014 ) , Maracay , Venezuela; the Ethics , Bioethics and Biodiversity Committee ( CEBioBio ) of the National Foundation for Science , Technology and Innovation ( FONACIT ) of the Ministry of Science , Technology and Innovation , Caracas , Venezuela; and by the Regional Health authorities of Aragua State ( CORPOSALUD Aragua ) . All adult participants signed written informed consent forms , and a parent or guardian of any child participant provided written informed consent on their behalf . Children between 8 and 17 years old provided written informed assent .
We described the general features , dengue knowledge and socio-economic characteristics of the study population and compared individuals interviewed with the adult versus the child questionnaire ( S1 and S2 Tables ) . The 105 interviewed individuals had a mean age of 40 years ( range: 18–87 years ) and were mostly females ( 86 . 7% ) . Children caregivers were older than those queried with the adult questionnaire ( mean age 44 vs . 35 years respectively; p<0 . 001 ) . This was expected , as those who cared for children were mainly mothers or grandmothers . Most of the interviewed individuals lived in Candelaria neighbourhood ( 68 . 6% ) . We were unable to complete the planned interviews in the other two neighbourhoods ( Cooperativa and Caña de Azúcar ) because of the previously mentioned anti-governmental protests . The majority of the 105 respondents completed secondary school ( 51 . 9% ) or higher education ( 31 . 7% ) and were housewives or domestic workers ( 53 . 8% ) . Those interviewed with the adult questionnaire had a higher education level and consisted of a higher proportion of students than parents/guardians of children . As in the rest of the country , the majority of the individuals professed a catholic religion ( 75 . 2% ) ( S1 Table ) . Most persons lived in households with five to six rooms ( 46 . 7% ) and households were mainly occupied by more than five inhabitants ( 71% ) . Households of children caregivers were more crowded than those of the ones interviewed with the adult questionnaire ( p = 0 . 003 ) , and had a lower monthly income ( p = 0 . 003 ) and socio-economic status ( p = 0 . 004 ) . There were no statistically significant differences between adults and parents/guardians with respect to the availability of public services , persons per household and the amount of household rooms ( S2 Table ) . The majority of interviewed individuals ( n = 103; 98 . 1% ) indicated that they had heard about dengue and showed good knowledge about dengue transmission . In response to the open question “how do you think people get infected by dengue ? ” , 95 . 2% ( n = 100 ) mentioned ‘the bite of a mosquito’ or similar . In response to the open question “what are the symptoms of dengue ? ” , nearly half ( n = 47; 44 . 8% ) mentioned up to three correct symptoms , the rest ( n = 58; 55 . 2% ) pointed out 4 or more , with a range of 0–8 symptoms . The overall knowledge score ranged from 1–9 correct answers ( S1 Table ) . The majority of the individuals ( n = 73; 69 . 5% ) reported to feel at risk of dengue . Almost all subjects ( n = 103; 98 . 1% ) also believed that people could die from dengue disease . Feeling at risk was equally reported when referred to children and adults ( n = 37; 72 . 5% vs . n = 36; 69 . 2%; p = 0 . 711 ) . Finally , 33 out of 103 respondents ( 32% ) mentioned that they/their child ( ren ) had dengue previously while one person did not know . Although a previous dengue infection was mentioned more frequently by parents/guardians of children than by adults ( 36 . 0% vs . 28 . 8% respectively ) the difference was not significant ( p = 0 . 440 ) . In order to understand the steps people would take in their search for health care , interviewees were confronted with the open questions: ‘what would you do if you/your child had fever’; and ‘what would you do if you think that you/your child had dengue’ . In the case of fever , most people chose to first treat fever at home ( n = 88; 83 . 8% ) versus only 12 ( 11 . 4% ) who mentioned that they would first seek medical help . In the case of dengue , the opposite was observed: most people would first visit a doctor ( n = 63; 60% ) , while nearly a third decided they would first treat dengue at home ( n = 31; 29 . 5%; p<0 . 001 ) . Measuring the temperature at home was mentioned by 35 ( 33 . 3% ) respondents in the case of fever and 29 ( 27 . 6% ) in suspected dengue . Less frequently proposed initial actions in the case of fever were ‘performing blood tests in a laboratory’ ( usually referring to a full blood count or platelet count tested at public or private laboratories ) , ‘inform my/the mother’ , and ‘to rest’ , while two adults decided they would take no action mentioning as reasons: ‘I will recover myself’ , ‘there is no need for visiting a doctor in case of fever’ and ‘I don’t like doctors’ . Other first intended actions mentioned with respect to dengue were ‘to perform blood tests’ , ‘to call a medical doctor’ , ‘to visit an alternative doctor ( not further specified ) ’ and ‘other’ ( inform mother , change the clothes of the child and use a mosquito net , call the dengue project staff , evaluate the disease ) . There were people who mentioned to do a blood test before going to the doctor . A tradeswoman and mother of a 9 years-old child explained this , referring to her daughter ( the quote is abbreviated ) : ‘When I go to the doctor , he will tell me to go to a laboratory to do blood tests . If I do a blood test before going to the doctor , this will save me the cost of one consultation . ’ This woman told us she would test for platelets when asked what she would do if she thought her daughter would have dengue . The first three actions individuals anticipated taking in the circumstance of fever or dengue , stratified by behaviour in the case of adults or children are presented as a flowchart in Fig 1 . Only the pathways that begin with either ‘home treatment’ or ‘visit medical doctor’ are shown , as these included 92 . 4% of all pathways . Proportions and analysis from now on in this article referring to Fig 1 , are based on the mentioned 92 . 4% sample . No differences were found when comparing the intended first actions and pathways between parents/guardians and adults in the case of fever , since most of them decided they would first treat fever at home , as depicted in the upper-left ( parents/guardians of children ) and upper-right panel ( adults ) of Fig 1 . Concerning dengue , differences were observed when comparing the intended first actions and pathways of children caregivers and adults , as can been seen in the lower left ( parents/guardians ) and lower right panel ( adults ) of Fig 1 . In the case of dengue , more adults than parents/guardians reported to first visit a doctor ( n = 36; 76 . 6% vs . n = 27; 57 . 4%; p = 0 . 048 ) , however children would have been taken earlier to the doctor than adults ( mean day chosen to visit a doctor: 1 . 3 ( n = 47 ) in children versus 1 . 5 ( n = 46 ) in adults; p = 0 . 108 ) ( Fig 1 ) . The combination of first treating at home and afterwards visiting a doctor , as a second step in the HSB pathway for suspected dengue , was reported more frequently by caregivers of children compared to adults ( n = 18; 38 . 3% vs . n = 9; 19 . 1%; p = 0 . 040 ) ( Fig 1 ) . To explore the association of determinants with the intended first action for both fever and dengue , we compared the characteristics of those who would first treat at home versus those who intended to first visit a doctor in univariate analysis ( S3 Table ) . The tested characteristics included child/adult-sample , age , sex , place of residence , education , occupation , religion , monthly income , socio-economic status , overall knowledge on dengue , reporting a previous dengue infection and risk perception . In the case of fever , those who would first visit a doctor were likely to be people with a lower educational level ( p = 0 . 069 ) , other variables did not show a significant association , probably due to the small sample size of those choosing to visit a doctor ( n = 12 ) . In relation to suspected dengue , the participants who chose to treat firstly at home consisted of higher proportions of people living in Caña de Azúcar ( p = 0 . 058 ) and caregivers of children ( p = 0 . 048 ) . All other tested variables did not show a significant association with any of the intended actions ( S3 Table ) . Table 1 shows the final multivariate models of factors that remained independently associated with visiting a doctor as the first intended action ( instead of choosing home treatment ) for either fever or suspected dengue . Respondents with a lower level of education were more likely to seek medical help as their first action in the case of fever , this relation was nearly significant . In the case of suspected dengue , individuals who referred having had a dengue infection in the past preferred to first treat dengue at home instead of going to a doctor firstly . Moreover , feeling at risk of dengue infection , not living in Caña de Azúcar neighbourhood and being an adult ( as opposed to a child caregiver ) in the case of suspected dengue were directly associated with the intention to first seek medical help ( Table 1 ) . In order to determine whether the intention to first treat at home would make people choose to go later to a doctor , we compared the day of seeking medical care from those who would first treat at home with those who would go firstly to the doctor . Those who intended to first treat at home reported a significant delay in their intentions to seek medical help versus those who intended to go to a doctor in the case of fever ( mean day = 1 . 93 vs . mean day = 1 . 50; p = 0 . 039 ) but not in the case of dengue ( mean day = 1 . 42 vs . mean day = 1 . 17; p = 0 . 098 ) . Overall , 90 ( 85 . 7% ) individuals stated they would treat fever at home at any time in their health seeking decision process while only 38 ( 36 . 2% ) would take this decision in the case of suspected dengue infection ( p<0 . 001 ) . Paracetamol ( an antipyretic ) was the most commonly chosen home treatment overall while taking a cold bath/shower and oral rehydration were the second most common types of home treatments in the case of fever and dengue , respectively ( Fig 2 ) . Within the people who reported to treat fever at home , most would use paracetamol ( n = 86 , 95 . 6% ) to lower the temperature , followed by a cold bath/shower ( n = 31 , 34 . 4% ) , oral rehydration ( n = 8; 8 . 9% ) , body sponging with a wet compress or sponge ( n = 7; 7 . 8% ) and other ways of home treatment ( n = 7; 7 . 8% ) such as rubbing the body with alcohol/cream , taking aspirin , other medication or rest . Paracetamol was also the most common choice among those who would treat dengue at home ( n = 30; 78 . 9% ) , however , the use of oral rehydration ( n = 9; 23 . 7% ) was cited more frequently than in the case of fever , opposite to the use of a cold bath/shower ( n = 7; 18 . 4% ) . Other ( n = 2; 5 . 3% ) home treatments for dengue included body sponging or rest . Combinations of home treatment for fever and dengue stratified by child or adult are shown in Fig 2 . The different choices for home treatment showed no significant differences between children and adults . If people reported to seek medical help when they/their child would have fever or suspected dengue , they were subsequently asked on which day after onset of first symptoms they would visit the doctor . Most parents/guardians of children and adults would seek medical help on day 2 after fever onset , but when dengue was suspected most people would go on day one to the doctor ( Fig 3 ) . Parents/guardians would take their children earlier to the doctor in case of dengue than in case of fever ( mean: 1 . 30 days vs . 1 . 78 days; p<0 . 001 ) . Referred to adults the mean reported day was 1 . 47 in case of dengue and 1 . 96 in case of fever ( p<0 . 001 ) . Although children would visit the doctor earlier than adults , this difference was neither significant in the case of fever ( p = 0 . 206 ) nor for dengue ( p = 0 . 162 ) . The most frequently reported reasons that prompted individuals to visit a doctor in suspected dengue were the appearance of new symptoms ( n = 81; 77 . 1% ) , the rise of body temperature ( n = 78; 74 . 3% ) and the persistence of fever ( n = 35; 33 . 3% ) . Almost 10% ( n = 10 ) of the people stated ‘dengue is a severe disease’ as a reason to seek medical help . The most frequent symptoms mentioned were headache ( n = 36; 34 . 3% ) , corporal pain ( n = 32; 30 . 5% ) and weakness ( n = 27; 25 . 7% ) . Other typical dengue symptoms were mentioned less frequently: rash ( n = 18; 17 . 1% ) , vomiting ( n = 11; 10 . 5% ) , eye pain ( n = 9; 8 . 6% ) and muscle pain ( n = 1; 1 . 0% ) . Moreover , warning symptoms were not frequently indicated as reasons to seek medical help: vomiting ( n = 11; 10 . 5% ) , bleeding ( n = 9; 8 . 6% ) , abdominal pain ( n = 6; 5 . 7% ) . The mean temperature referred by interviewees was 39 . 4°C ( range: 38 . 0°C—42 . 0°C ) while persistence of fever ranged between one to four days with a mean of two days . There were no significant differences when comparing the reasons to seek medical care in suspected dengue between children and adults .
Due to the political unrest that took place during the study period , data collection was not completed for all three neighbourhoods and the majority of the interviewees resided in Candelaria neighbourhood . However , this population was considered to be representative of most urban areas from Maracay city minimising selection bias . Education , monthly income and socio-economic status were significantly lower in the child sample , nonetheless these variables were controlled for in the multivariate analysis . The difference in socio-demographic and socio-economic characteristics between the adults and parents/guardians of children can be attributed to the fact that almost half of the children in Aragua State , Venezuela , are part of a household run by a single mother , who have generally a lower degree of education and a lower income [30] . A strength of the study was that data was collected from a well characterised cohort study population of which socio-economic and epidemiological data was available . Moreover , contrary to hospital-based studies , our study design made us able to include people who would avoid attending health centres , thus obtaining insight in their intended HSB . Furthermore , people were interviewed at home , providing a safe and confident environment . Finally , the analysis was strengthened with the conceptualisation that included aspects of the HBM . For future studies , we recommend to apply the conceptualisation at the stage of data collection . In the current study , we were able to describe intended HSB in the case of dengue and fever at community level . Our results suggest that for the people who intend to seek medical care in the case of a dengue infection , self-diagnosis might be an obstacle . The differences found in HSB between fever and dengue imply that fever does not increase the perceptions of susceptibility to dengue . As mentioned before , a delay in care-seeking is associated with a higher mortality and complications during dengue disease [1 , 12] . Therefore , the early intended medical care-seeking in the case of a suspected dengue infection shown in this study suggests a possible improvement of HSB and prognosis if an algorithm or tool can be designed to diagnose dengue at home . In addition , we found that those who previously had a dengue infection were more likely to treat a next infection firstly at home . Since this group has a higher chance of developing a severe dengue disease , efforts have to be made to promote prompt health centre attendance in this group . In this , raising awareness and risk perception of dengue by media coverage and information at health centres may improve this favourable behaviour . Comparing the results of the current study ( HSB intentions ) with those of the actual HSB taken by people , such as in health centre-based studies , could reveal other possible barriers for achieving the intended HSB . More community and health centre-based studies should be performed to achieve a wider view and stronger conclusions on HSB for people exposed to dengue in the Americas .
|
The rapid spread of dengue in the last decade has brought this disease higher in the public health agenda worldwide . Dengue is transmitted by day-biting mosquitoes and typically presents with sudden fever and flu-like symptoms . In a proportion of cases , severe disease can ensue which if not promptly treated can be lethal . In Venezuela , the frequency of dengue epidemics and severe disease is on the increase . Delay in care-seeking has been related to the development of serious disease complications . In this study , we describe the health-related behaviour intentions people have when they face a dengue infection , to find ways to promote prompt medical care attendance . Few studies of this kind exist in the Americas . We found that in case of fever alone , people would mostly prefer to first treat the disease at home while in the case of suspected dengue they would choose to promptly visit a doctor . These intentions are encouraging . However , since people often do not realize they have dengue , they may not perform their planned behaviour . Moreover , we found several factors that made people choose to visit promptly a doctor: being an adult ( as opposed to caring for a child as a parent/guardian ) , place of residence , feeling at risk and not having had a previous dengue infection . Therefore , we suggest that raising risk perception and the knowledge on dengue symptoms and transmission route could improve earlier health centre attendance lowering the risk of clinical complications leading to severe dengue and death .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Health Seeking Behaviour and Treatment Intentions of Dengue and Fever: A Household Survey of Children and Adults in Venezuela
|
The cystic fibrosis ( CF ) lung microbiome has been studied in children and adults; however , little is known about its relationship to early disease progression . To better understand the relationship between the lung microbiome and early respiratory disease , we characterized the lower airways microbiome using bronchoalveolar lavage ( BAL ) samples obtained from clinically stable CF infants and preschoolers who underwent bronchoscopy and chest computed tomography ( CT ) . Cross-sectional samples suggested a progression of the lower airways microbiome with age , beginning with relatively sterile airways in infancy . By age two , bacterial sequences typically associated with the oral cavity dominated lower airways samples in many CF subjects . The presence of an oral-like lower airways microbiome correlated with a significant increase in bacterial density and inflammation . These early changes occurred in many patients , despite the use of antibiotic prophylaxis in our cohort during the first two years of life . The majority of CF subjects older than four harbored a pathogen dominated airway microbiome , which was associated with a further increase in inflammation and the onset of structural lung disease , despite a negligible increase in bacterial density compared to younger patients with an oral-like airway microbiome . Our findings suggest that changes within the CF lower airways microbiome occur during the first years of life and that distinct microbial signatures are associated with the progression of early CF lung disease .
Cystic fibrosis ( CF ) is a multisystem genetic disease in which pulmonary manifestations account for the majority of morbidity and mortality . CF lung disease is characterized by thickened airway secretions , bacterial infection , and neutrophil dominated inflammation that leads to progressive airway destruction ( bronchiectasis ) and , ultimately , respiratory failure [1] . Traditional pathogens associated with CF lung disease include Staphylococcus aureus and Haemophilus influenzae in the first years of life , followed by an increasing prevalence of Pseudomonas aeruginosa in older children and adults [2] . Sensitive molecular based ( 16S rRNA gene sequencing ) microbiome analysis of respiratory secretions from children and adults with CF suggest that infection is polymicrobial and often includes both traditional CF pathogens and aerobic and anaerobic bacteria typically found in the oral cavity [3–8] . Extended culture methods have confirmed that bacteria typically found in the oral cavity are present and viable in CF respiratory secretions , with densities similar to those of pathogens [9 , 10] and higher than that seen in samples from healthy volunteers [11] . Oral bacteria have also been found in distal areas of the lung from a young CF patient at time of lobectomy [12] . However , contamination of lower airways samples by oropharyngeal secretions during collection remains a significant concern in airway microbiome research [13] . Consistent with this concern , it has been shown that oral bacteria are abundant in sputum and throat swabs from end-stage CF patients immediately prior to lung transplant , but airway samples obtained directly from the explanted lungs immediately after transplant are dominated almost exclusively by traditional CF pathogens [14] . Carefully controlled bronchoscopy studies indicate that the lower airways microbiome of healthy volunteers is similar in composition , albeit significantly lower in density , to that of the oropharynx [15–18] . The lower airways microbiome in healthy individuals is likely derived from normal microaspiration of upper airways secretions and not contamination during bronchoscopy [13 , 19 , 20] . In healthy individuals , oropharyngeal bacteria are presumed to be transient in the lower airways and their detection is unlikely to represent true colonization . As seen in microbiome studies of healthy porcine lungs , much of the bacterial DNA recovered by bronchoscopy is DNase sensitive , indicating that it is largely derived from bacteria that have been killed by lung defenses [21] . Increased DNA from oral bacteria in the lower airways , during disease , may reflect increased residence time due to defective clearance or diminished innate immune function rather than colonization . It has been proposed that exposure of the lower airways to anaerobes and/or oral bacteria , even if transient , is likely to influence lung physiology , local metabolites and mucosal immune homeostasis [20 , 22 , 23] . In several countries ( United Kingdom , Germany , and most CF Centers in Australia ) antibiotic prophylaxis is given to children with CF until they are two years of age [24] , while other European countries treat any bacteria detected in cough-swabs during routine CF clinic visits [25] . In the United States prophylactic antibiotic therapy for infants and young children with CF is not recommended [26]; however , a recent clinical trial in the US and Canada showed that children less than 6 years of age spent an average of two months per year on antibiotic therapy for increased respiratory symptoms [27] . Airway inflammation in CF can be detected within the first few months to years of life , despite the absence of clinically diagnosed infection [28 , 29] . Furthermore , increased inflammation and structural lung disease occurs in children with CF despite the use of prophylactic antibiotics [30–33] . The underlying cause of early airway inflammation in CF remains controversial and may reflect undiagnosed infection [28 , 29 , 34] , altered immune function [35 , 36] , or a response to mucus obstruction [37] . To better understand the microenvironment in very young , clinically stable , CF subjects , we applied sensitive molecular detection methods to characterize the microbiome of the lower airways using cross-sectional BAL samples obtained from infants and preschoolers who underwent bronchoscopy as part of the Australian Respiratory Early Surveillance Team for Cystic Fibrosis ( AREST CF ) study . The early CF lower airways microbiome was analyzed in relation to antibiotic prophylaxis , BAL markers of inflammation and structural lung disease , measured by chest computed tomography ( CT ) , to elucidate the evolution of the lower airways microbiome with respect to disease progression .
To examine the early lower airways microbiome in CF , we obtained BAL samples collected during annual AREST CF study bronchoscopies from 46 CF subjects , ranging in age from approximately 3 . 5 months to five years with a median age of 1 . 95 years and an interquartile range ( IQR ) of 1 . 13–4 . 06 years . All subjects were diagnosed with CF through newborn screening . Patients were intubated during CT and bronchoscopy to control breathing and avoid aspiration . Per Australian clinical care standards , all children with CF are prescribed amoxicillin-clavulanic acid as antibiotic prophylaxis during the first two years of life . Thirty samples were obtained from subjects aged two years or younger . Seventeen samples showed clinically defined infection ( i . e . , BAL cultures showed a density of ≥105 colony forming units/ml of a recognized pathogen ) . Twenty-three subjects were homozygous and 22 subjects were heterozygous for the F508del CFTR mutation . One subject had two other CFTR mutations . All subjects were seen routinely by a CF nutritionist and recommended a high fat diet with enzyme supplementation . Breastfeeding status was not collected . For 25 patients , we processed two separate aliquots of pooled BAL collected from the right middle lobe ( RML ) for methods development . A single RML-derived BAL sample was studied from an additional 21 patients . Seven of the 46 subjects in this study had a second longitudinal BAL sample collected from the RML four to 22 months after the initial sampling . Longitudinal samples were included in observational analyses , while only a single sample per patient was considered in statistical analyses ( see Methods ) . Clinical and study related data for all patients/samples is provided as Supporting Information ( S1 Appendix ) . To control for background signal introduced during patient sampling , a saline wash was collected from two study associated bronchoscopes prior to BAL; bronchoscope washes were not available for all samples . Additionally , an aliquot of sterile water was included as a process control at the DNA extraction step . All patient and control samples were analyzed by bacterial 16S rRNA gene sequencing and quantitative polymerase chain reaction ( qPCR ) to determine bacterial identity and density , respectively . DNA was extracted from all samples as previously reported for similar studies [14 , 15] . Given that infant BAL is likely to have low biomass , we randomly selected a subset of patient samples for protocol development , using the method of Lundberg et al . as a starting point [38] . Empirical testing indicated that 25 cycles of initial amplification of the V4 16S rRNA gene region followed by 20 cycles of barcoding was necessary to generate sufficient 16S rRNA gene amplicons for sequencing . All samples , including controls were individually barcoded , pooled and sequenced simultaneously . After sequence processing and quality control , a total of 7 . 5 million sequences were classified into 423 Operational Taxonomic Units ( OTUs ) . Given the potential for PCR bias in low biomass samples , we assessed the reproducibility of our methods by comparing the 16S rRNA gene sequence data for all 25 duplicate BAL aliquots ( three were longitudinal ) , which matched the age distribution of the larger cohort ( median , 1 . 86 years; IQR , 1 . 11–3 . 97 ) . For analysis , these replicate samples were partitioned from the main sample set and rarefied to 6 , 000 sequences to retain all samples and represent their sequencing depth evenly . For the 25 duplicate BAL aliquots , the median Pearson product-moment correlation coefficient ( PPMCC , r ) between replicates was 0 . 98 ( IQR , 0 . 81–1 . 00 ) , at the OTU level . Community composition was highly similar between most replicates when OTUs were binned to both phylum ( Fig 1A ) and the lowest identifiable taxonomic level ( Fig 1B ) , where the median PPMCC was 0 . 98 ( IQR , 0 . 85–1 . 00 ) for the latter . Non-metric multi-dimensional scaling ( NMDS ) analysis of the replicates showed that patient samples separated into discrete groups ( S1 Fig ) . Those replicates with the lowest correlative values generally co-localized to the same group ( Fig 1B , S1 Fig ) and were subsequently determined to contain low biomass ( S1 Appendix ) and background sequences ( see below ) . Comparison of the replicate samples suggested that our methodology was sufficiently robust , and that single BAL samples were likely to provide a reasonable description of the lower airways microbiome in our cohort . Given the general agreement of the replicate samples , sequence data were pooled for each of the 25 duplicate BAL aliquots and rarefied to 11 , 000 sequences . Analysis of the combined aliquot data by NMDS confirmed that the samples separated into three discrete groups at the OTU level , designated as G1 , G2 and G3 ( Fig 2 ) . G1 was composed of samples with the lowest replicate correlative values ( median , 0 . 68; IQR , 0 . 44–0 . 87 ) , while samples with the highest replicate correlative values segregated to G2 ( median , 0 . 99; IQR , 0 . 98–1 . 00 ) and G3 ( median , 1 . 00; IQR , 1 . 00–1 . 00 ) . The groups also showed a trend with regard to patient age ( Fig 2 ) ; G1 including the youngest patients ( median , 1 . 08 years; IQR , 1 . 00–1 . 17 ) , while G2 ( median , 1 . 67 years; IQR , 1 . 20–1 . 96 ) and G3 ( median , 4 . 09 years; IQR , 4 . 02–4 . 31 ) included increasingly older patients . Shannon diversity indices for each group were similar between G1 and G2; however , G3 was significantly lower than G2 ( S2A Fig ) . The samples within G1 tended to have the largest differences in diversity scores between replicate pairs ( median , 0 . 52; IQR , 0 . 27–0 . 71 ) as opposed to G2 ( median , 0 . 20; IQR , 0 . 15–0 . 25 ) and G3 ( median , 0 . 09; IQR , 0 . 03–0 . 29 ) . Analyses of the two bronchoscope washes and a process control ( three background controls ) showed high correlations with each other at the OTU and genus level ( average r = 0 . 94 , r = 0 . 94 , respectively ) . As shown previously , the background signal originating from sample collection reagents and DNA extraction kits [39 , 40] can dominate extremely low biomass samples and provide a signature for samples that otherwise lack appreciable amounts of bacteria [41] . Approximately 90% of the sequences in the three background controls were sourced from OTUs assigned to families Enterobacteriaceae , Bradyrhizobiaceae , and Comamonadaceae ( S3A Fig ) . Previous sequencing studies have reported at least two of these taxa as contaminants derived from DNA extraction reagents [39 , 40] . To determine whether background sequences contributed to the observed BAL OTU groups , we compared the average sequence signature for the background controls to the average for each of the observed sample groups . At the OTU level , the average background signal correlated highly with G1 ( r = 0 . 81 ) and not with G2 or G3 ( r = 0 . 02 , r = -0 . 01 , respectively ) . Further , the average background signal grouped with G1 samples , by NMDS ( Fig 2 ) . OTUs within G1 were dominated by Bradyrhizobiaceae , Comamonadaceae and Enterobacteriaceae , similar to the background controls ( S3B Fig ) . In contrast , G2 was dominated by Streptococcus and G3 was dominated by Haemophilus and Moraxella ( S3B Fig ) . Based on this analysis , we conclude that two potential bacterial community types are present in our early CF cohort , represented by G2 and G3 , while G1 samples were largely explained by background signal and represent samples that lack an appreciable host-derived microbiome . To determine whether these groups were maintained in a larger sample set , we expanded our analysis to include an additional 28 BAL samples ( four of which were longitudinal ) for which only a single aliquot was available . In addition , the average background signal was treated as an independent sample and included for reference . NMDS analysis of the complete cohort , at the OTU level , showed that many of the additional samples distributed closely with the previously defined groups ( G1-G3 ) . A subset of samples appeared to contain unique OTU distributions or a mixture of OTUs from multiple groups ( S4 Fig ) . Additionally , the age distribution of the samples continued to exhibit a group associated trend ( S4 Fig ) . To identify the taxonomic associations that distinguish samples within our cohort , we conducted a principal component analysis ( PCA ) . Previous studies have demonstrated that low biomass airway microbial community structure can be adequately defined by considering only the most abundant community members [41–43] . Therefore , to reduce complexity in our dataset , we considered taxa that contributed ≥0 . 5% of the average relative abundance across all samples . Based on this cut-off , 23 taxa accounted for >95% of all sequence data . Principal component analysis of all samples indicated that these 23 taxa separated into three distinct associations or clusters , designated C1 , C2 and C3 . ( Fig 3 ) . Cluster type C1 was defined by seven taxa; most often associated with the environment or abundant in our background controls ( e . g . , Enterobacteriaceae , Comamonadaceae , and Bradyrhizobiaceae ) . Further , the average background signal grouped with the C1 dominated BAL samples ( Fig 3 , white circle ) , indicating that these samples lack an appreciable lung derived microbiome . C2 was primarily defined by ten taxa typically associated with the oral cavity ( e . g . , Streptococcus , Prevotella and Veillonella ) . C3 was represented by six taxa , including those recognized clinically as pathogens in CF ( e . g . , Haemophilus , Staphylococcus , Moraxella and Pseudomonas ) . Shannon diversity for C1 and C2 samples were similar , while C3 values were significantly lower ( S2B Fig ) . To further examine and visualize the relationships between the members of these cluster types , we generated a correlation matrix for the 23 taxa used in the PCA , across all BAL samples ( S5 Fig ) . As expected , there was a positive correlation between taxa within C1 ( average r = 0 . 39±0 . 12 ) and C2 ( average r = 0 . 43±0 . 06 ) . However , taxa in C1 and C2 negatively correlated with each other ( average r = -0 . 29±0 . 10 ) , suggesting that the C1 and C2 clusters of bacteria tended to be mutually exclusive within our BAL samples . In contrast , taxa within C3 did not correlate with each other ( average r = -0 . 02±0 . 03 ) or with taxa from C1 or C2 ( average r = -0 . 08±0 . 11 and -0 . 10±0 . 05 , respectively ) , indicating that members of cluster type C3 generally occur independently of each other consistent with the lower diversity ascribed to this cluster . Given that >95% of sequences could be assigned to three associative clusters , we examined the BAL samples based on the relative proportion of each cluster ( Fig 4 ) . When viewed in this way , it was clear that some samples contained a mixture of cluster types , which helped describe the overall lower airways microbiome structure of the study cohort when observed through NMDS ( S6 Fig ) . C1 dominated samples were most prevalent in the youngest patients ( <1 year ) , whereas the C3 signature was generally associated with older subjects ( >4 years ) ( Fig 4 ) . The C2 BAL cluster type was most prevalent in the intermediate age group . While cross-sectional , these results suggest that the dominant cluster type may change in a temporal or age-associated manner . To test this hypothesis , each patient sample was defined as C1 , C2 or C3 according to its dominant cluster type ( S1 Appendix ) . Comparison of patient age based on these cluster designations showed that C1 dominated patients were generally younger , but not significantly different from C2 . In contrast , C3 designated samples came from patients that were significantly older than both C1 and C2 patients ( Fig 5A ) . Samples assigned to C1 were associated with background signal , suggesting that they harboured extremely low levels of airways derived bacterial DNA . To determine if bacterial burden differed between samples based on the assigned dominant cluster type , we used qPCR to assess absolute bacterial 16S rRNA gene copies in each sample ( Fig 4 ) . Groupwise comparisons showed that C2 and C3 dominated samples were not different from each other , but had significantly greater bacterial densities than C1 ( Fig 5B ) , which was similar to background ( Fig 4 ) . This further supports the conclusion that C1 dominated samples lack an appreciable lung-derived bacterial microbiome . As expected , there was a positive correlation between age and 16S rRNA gene copy number ( r = 0 . 32; p≤0 . 05 ) . Cluster assignments were further supported by clinical culture results , which showed that 10 of 12 subjects in C3 were infected with a known CF pathogen compared to 6 of 21 in C2 and 1 of 20 in C1 ( S1 Appendix ) . We next evaluated the relevance of the dominant BAL cluster types to disease progression using BAL markers of inflammation and CT-defined structural lung disease as measures of disease severity ( Fig 5C–5G , S1 Table ) . Patients with BAL samples defined as C1 had the lowest inflammation ( total cell counts , total neutrophils and IL-8 ) and minimal to no detectible structural lung disease ( bronchial wall thickening and bronchiectasis ) based on CT scan . Patients with samples identified as C2 showed intermediate inflammation , with significantly higher total cell counts and total neutrophils compared to those designated as C1 , but showed minimal structural change to the airways . While C2 samples were dominated by oral associated bacterial sequences , they often contained a small proportion of pathogen derived ( C3 ) sequences ( Fig 4 ) . However , the presence of pathogen sequences in C2 designated samples did not account for the increase in inflammation ( S7 Fig ) in this patient group . Those patients with samples belonging to C3 exhibited further increases in both , inflammation and structural lung disease ( Fig 5 , S1 Table ) . To account for additional factors , including study site and antibiotic prophylaxis , we conducted multivariate analyses of the clinical data . When adjusted for these factors , the data indicated significantly higher BAL inflammation in patients with samples identified as C2 and C3 compared to C1 ( S1 Table ) , and increased structural lung disease ( bronchiectasis ) in C3 designated patients compared to C1 . Taken together , these results suggest that increasing disease severity correlates with changes in the lung microbiome . To evaluate the specific contribution of oral flora and conventional CF pathogens , as distinct groups , to disease severity and progression , we used multiple regression analysis ( S2 Table ) . When corrected for the presence of conventional pathogens , the relative abundance of oral-associated taxa ( as a group ) was significantly associated with increased bacterial 16S rDNA copy number , total cell counts and total neutrophils . These results demonstrate that the presence of an oral-like lower airways community is directly associated with early inflammation . In contrast , increasing relative abundance of conventional pathogens ( as a group ) was significantly associated with increased bacterial 16S rDNA copy number , measures of inflammation and structural lung disease ( bronchial wall thickening and bronchiectasis scores ) . Further , the increasing relative abundance of both oral taxa and pathogens were significantly associated with measures of early inflammation when regression analyses were controlled for patient age ( S2 Table ) . While the vast majority of samples in this study were cross-sectional , we evaluated seven subjects who had longitudinal BAL samples to test for temporal related changes in bacterial cluster type . Samples from five of the seven subjects did not show a change in cluster type over time . However , samples from two subjects showed a changed from an initial background dominated cluster type ( C1 ) to an oral dominated cluster type ( C2 ) ( S4 , S6 and S8 Figs ) . Interestingly , the single pathogen dominated longitudinal pair showed a shift from Haemophilus to Moraxella during the sampling interval ( S8 Fig ) . Despite this genus level shift , the pathogen cluster type ( C3 ) was maintained . While the number of longitudinal samples available for this study was limited , our analysis suggests that the relative composition of taxa within a given cluster type may be dynamic over time , but cluster typing designations are relatively stable . When cluster type did change within this subset of patients , it resulted in progression from C1 to C2 .
This study describes early changes in the CF lower airways microbiome in a unique study population , namely infants and preschool children who underwent bronchoscopy and CT scan at times of clinical stability . This young and clinically stable population showed absence of an appreciable lower airways infection in a subset of samples derived from the youngest subjects . Given the inherent low biomass in BAL samples from our cohort , extensive PCR amplification was needed to enhance sensitivity , which increased the risk of detecting background contamination [40] . Here , the background signal allowed for the differentiation and comparison of samples that lacked appreciable host derived bacteria , from those in which bacteria were present in appreciable quantities . The microbial signal present in BAL from the C1 subjects closely resembled the background detected in bronchoscope washes and processing reagents . The predominance of background sequences and low detectible bacterial biomass by qPCR ( approximately 100 bacterial 16S rRNA gene copies/ml BAL on average ) enhances our confidence that C1 designated samples came from patients that were not bacterially infected . This was consistent with clinical microbiology results , which showed that only one of the twenty C1 samples was culture positive for a pathogen . Importantly , the relative abundance of background sequences became negligible in the C2 and C3 samples , where non-background taxa dominated and bacterial biomass was increased by 3–4 orders of magnitude . Comparison of duplicate aliquots of the same BAL samples showed that our methodology was reproducible and that PCR bias was negligible . As expected , the greatest variability between replicates was seen in C1 samples that lacked appreciable patient derived bacterial DNA . Interestingly , we found that bacterial sequences detected in the lower airways samples of these CF infants and young children separated into three distinct cluster types . Further evaluation indicated that samples from our C1 group lacked a true lower airways bacterial community and represented an uninfected state . C2 samples harbored an actual microbial community resembling that found in the upper airway or oral cavity . We hypothesize that these samples represent the initial acquisition of a lower airways microbiome . Finally , we found that older subjects often harbored a lower airway bacterial community dominated by one or a few traditional CF pathogens . We hypothesize that the increased abundance of bacteria typically found in the oral cavity in BAL reflects repeated micro-aspiration . Despite aspiration typically being more prevalent in infants less than one year of age , and bacterial communities being established in the oral cavity in the first few months of life [44 , 45] , subjects with BAL samples dominated by oral microbiota were mostly >12 months old in this study . Thus , we hypothesize that additional events early in life are necessary to establish lower airway niches capable of supporting the accumulation and persistence of oral microbes ( e . g . , decreased mucociliary clearance , increased mucus accumulation , and an anaerobic microenvironment ) . Given that bacterial density in BAL was similar in samples dominated by oral bacteria and pathogens , it seems possible that orally derived microbes may colonize or infect the lower airways; however , our study was not designed to distinguish airway colonization or infection from reduced clearance of aspirated bacteria . The oral community in these BAL samples included taxa identified in studies of older CF subjects by both molecular-based [3 , 4 , 8] and culture-based detection with densities approximating those of pathogens [11] . The role that these aerobic and anaerobic oral bacteria play in CF pathogenesis remains a subject of discussion . Recent functional studies have demonstrated that members of the oral microbiota may play an important role in conditioning the mucin rich lower airways environment for subsequent colonization by pathogens [22] . While many CF pathogens are incapable of efficiently metabolizing intact mucins for growth , consortia of oral microbes can use mucin carbohydrates as a primary nutrient source [22 , 46] . Metabolites derived from degradation of mucins by these oral-derived consortia , support growth of traditional pathogens [22] . In addition , short chain fatty acids released by fermentation also contribute to inflammation [47] . As such , the oral dominated community type , detected in our study , may drive nutritional changes and inflammation in the lower airways environment , providing a potential mechanistic explanation for the increase in BAL markers of inflammation and the microbial succession seen in our young CF cohort . Those subjects with BAL samples harbouring recognized pathogens had the highest markers of inflammation and overt structural lung disease . When present , pathogen sequences were at a higher relative abundance than other taxa , despite the BAL not having significantly greater microbial density than oral bacteria dominated samples . This dominance of pathogens is consistent with previous reports of decreased microbial diversity in oropharyngeal ( OP ) swabs across a wide age range [3 , 48 , 49] . Sputum based studies typically showed a progressive decrease in diversity with increasing disease [4 , 50 , 51] . In older patients , these findings may reflect increased antibiotic exposure [52] and/or once acquired , pathogen competition [53] . Here , progression of the airway microbiome and disease markers was heterogeneous , but generally changed with age . Multivariate analyses adjusted for age , indicated that the association of the pathogen dominated community type with increased neutrophil counts and bronchiectasis remained significant . Although contamination of BAL with oral secretions can occur during bronchoscopy , this was minimized by placement of an endotracheal tube prior to the procedure . Importantly , contamination during the procedure does not explain the age dependent appearance of distinct community types , or the presence of lower airway inflammation with oral microbiota . We acknowledge that we did not have access to healthy non-CF infant control BAL samples to assess the normal lower airways microbiome or measure baseline inflammatory markers; however , given the invasive nature of bronchoscopy , collection of such samples would not be ethical . Further , the focus of this study was to elucidate the development of a lower airway microbiome in children with CF . All children in this study were prescribed amoxicillin-clavulanic acid during the first two years of life , which could have affected our results , or delayed the onset of lower airways acquisition of oral microbes . Our multi-variate analyses corrected for this; however , we cannot accurately determine or predict the effect of early antibiotics has on presence of the lower airway microbiome in general . Antibiotics are prescribed frequently and early to infants and children with CF during times of increased respiratory symptoms [27] and antibiotic prophylaxis is recommended in young children with CF in several countries [24] . However , in the face of antibiotic prophylaxis a transition from background to an oral-like lower airways microbiome occurred , which may imply that bacteria are able to persist or accumulate in lower airways niches ( e . g . , mucus ) despite antibiotic exposure . Further , adjustment for antibiotic prophylaxis in the multivariate analyses showed progressive disease in those dominated by oral bacteria at an age while children were still on prophylaxis . True longitudinal studies of BAL microbiome in infants and children are difficult to perform but several groups have taken approaches to assess the development of the airway colonization in CF children . Hoen et al . examined the development of the gastrointestinal and respiratory microbiome in infants with CF , where respiratory samples were obtained by OP swab [54] . They showed a core microbial community spanning oral and intestinal samples that included genera detected in our study . For the gut microbiota , a predictive change occurred prior to onset of P . aeruginosa detection . However , as shown by clinical microbiology , the predictive value of OP swabs to BAL derived lower airways samples is modest [55 , 56] . In our study , such comparisons of oral and/or stool microbiome to BAL microbiome was not possible as the AREST CF protocol does not include upper airway samples or stool collection . A more recent study compared the microbiome of OP and nasopharyngeal ( NP ) swabs to BAL in children with different underlying airway diseases . The study found that microbial community composition differed significantly between the sample types [57] . Nonetheless , the underlying microbial communities detected were predictive of disease state ( i . e . , prolonged cough , non-CF bronchiectasis , and children without lower airway infection ) . Longitudinal comparisons of nasal microbiota have revealed that differences emerged within the first year of life in CF compared to healthy infants [58] . Comparison of the NP microbiome in infants with CF to healthy infants showed increasing separation of the microbial composition ( e . g . , increased Staphylococcus OTUs in CF ) between the groups during the first six months of life [59] . Our data using BAL sampling in all subjects supports the premise that CF infants are initially uninfected . With time , our subjects exhibited a progressive increase in the density of oral bacteria in the lower airways , which ultimately changed to a pathogen dominated environment . This conclusion is consistent with data from a small subset of patients in our study that had longitudinally collected BAL . Five of seven patients maintained the same microbial community type between sampling intervals ( 3 . 6 to 24 months ) , while two subjects progressed from an uninfected or background cluster type to an oral bacteria dominated community over time . This study is unique in that it utilized lower airways samples obtained by bronchoscopy through an endotracheal tube from clinically stable subjects with CF ranging from infancy to preschool age . The study revealed that most CF subjects less than one year old had negligible bacterial densities in their BAL . Subsequently , a change in composition was seen with predominance of oral bacterial sequences in the 1–2 year olds and increasing presence of pathogens in samples from 3–5 year old subjects . The acquisition of an oral bacteria dominated lower airway microbiome and the succession to a pathogen dominated microbiome were each associated with increased markers of disease . Based on our analyses , we conclude that stratification of patients based on BAL cluster type was highly indicative of disease status . Testing the hypotheses of microbial community succession , the relevance of oral aspiration , and mucus plugging as a contributor to disease progression will require a larger longitudinal sample set .
AREST CF conducts annual bronchoscopy and chest CT scans on infants and children diagnosed with CF when clinically stable [60] . In this study we included BAL samples from children who had undergone bronchoscopy and CT between 2011 and 2013 . Subjects were enrolled in the CF clinics at Princess Margaret Hospital for Children , Perth , Australia and the Royal Children’s Hospital , Melbourne , Australia . Antibiotic prophylaxis with amoxicillin-clavulanic acid ( 15 mg/kg/day ) is prescribed during the first two years of life to all children with CF in Perth and Melbourne . Subjects / BAL samples in this study were stratified to include a range from the earliest BAL to 5 years with and without infection on BAL diagnosed by clinical microbiology . All BAL samples used in this study were collected under AREST CF protocols at Princess Margaret Hospital for Children , Perth , Australia and the Royal Children’s Hospital , Melbourne , Australia . AREST CF protocols were approved by Ethics Committees at both institutions . All samples were anonymized and stored for research purposes . The parents of participating children were informed of risks and given time to agree or decline voluntary participation prior to the start of all research procedures . Written parental consent was obtained for all subjects and included approval to share samples with other studies . The use of anonymized AREST CF BAL samples in this study was approved by the University of North Carolina at Chapel Hill office of Human Research Ethics . Bronchoscopy and CT scan procedures were performed as previously described [60] . Briefly , intravenous general anaesthesia with intubation was used to control breathing for the CT scan and to avoid contamination of the bronchoscope . BAL was performed with three aliquots of 1 ml/kg normal saline instilled into and aspirated from the right middle lobe ( RML ) and one aliquot into the most diseased lobe per CT-scan . The first aliquot from the RML was collected in a separate suction trap and sent to clinical microbiology lab for culture . The pooled 2nd and 3rd aliquots were collected into a single suction trap and stored frozen at -80°C in separate 1ml aliquots . These aliquots were subsequently used for assessment of inflammatory markers and microbiome analysis . Volumetric CT scans were used for inspiratory assessment and either a 3-slice protocol at end expiration or volumetric expiratory assessment . The previously used semi-quantitative CF-specific CT scoring system measures gas trapping , bronchial wall thickening and bronchiectasis [30 , 61] . Markers of lung inflammation in BAL included total and differential cell counts , and IL-8 measured by ELISA as done previously [62 , 63] . The V4 region of the bacterial 16S rRNA gene was targeted for sequencing in a two-step preparation , using modified universal primers adapted from Lundberg et al . [38] . Bead-cleaned , equimolar concentrations of amplicons ( approximately 450 base pairs in size ) were sequenced on an Illumina MiSeq using a V2 paired-end 500 cycle kit . Raw sequence data was deposited in the European Nucleotide Archive ( Study ID PRJEB13657 ) . Read and data processing was performed in QIIME 1 . 8 . 0 and R version 3 . 2 . 3 . Specific details of quantification , sequencing , and analysis are provided ( S1 File ) . Clinical data were archived prospectively in the AREST CF database . Data are reported as median and interquartile ( 25–75% ) range using inflammatory markers and CT scores as continuous outcomes . To account for multiple sampling , patients with longitudinal samples had either time point 1 or time point 2 randomly selected to be included in all statistical comparisons . This random selection was performed once and maintained throughout all analyses ( see S1 Appendix ) . Group comparisons were performed using Tukey’s HSD with a significance threshold of 0 . 05 . Multiple regression analysis was performed in R version 3 . 3 . 3 . Multivariate analyses used linear mixed effects models adjusted for age , antibiotic prophylaxis , and study site with significance at 0 . 05 . Analyses were performed using Stata 13 . 0 ( StataCorp LP ) and JMP Pro 12 . 0 . 1 Software ( SAS ) .
|
CF lung disease is characterized by persistent airway infection by complex microbial communities . These communities often consist of pathogens and endogenous microbes typically associated with the oral cavity . The development of these complex communities and their relationship to CF lung disease progression is unclear . To understand the evolution of the CF lower airways microbiome , we applied sensitive molecular detection methods to characterize the bacterial DNA sequences in bronchoalveolar lavage ( BAL ) samples obtained from clinically stable infants and preschoolers who underwent bronchoscopy . Our findings demonstrate that CF infants have relatively sterile lower airways with a progressive shift to a microbiome dominated by aerobic and anaerobic bacterial species commonly associated with the oral cavity . This initial acquisition of a lower airways microbiome was associated with a significant increase in bacterial burden and increased airway inflammation . Transition from an oral dominated to a pathogen dominated lower airways microbiome correlated with a further increase in inflammation and the onset of structural disease despite a negligible increase in bacterial density . Our findings suggest that oral microbes may play an important role in early CF airway disease and could potentially predispose subjects to subsequent infection by pathogens .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"inflammatory",
"diseases",
"children",
"medicine",
"and",
"health",
"sciences",
"microbiome",
"pathology",
"and",
"laboratory",
"medicine",
"diagnostic",
"radiology",
"pathogens",
"genetic",
"diseases",
"microbiology",
"fibrosis",
"neuroscience",
"pulmonology",
"preventive",
"medicine",
"age",
"groups",
"developmental",
"biology",
"cystic",
"fibrosis",
"infants",
"bacterial",
"pathogens",
"microbial",
"genomics",
"families",
"autosomal",
"recessive",
"diseases",
"research",
"and",
"analysis",
"methods",
"neuroimaging",
"public",
"and",
"occupational",
"health",
"antibiotic",
"prophylaxis",
"imaging",
"techniques",
"medical",
"microbiology",
"microbial",
"pathogens",
"tomography",
"computed",
"axial",
"tomography",
"clinical",
"genetics",
"people",
"and",
"places",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"prophylaxis",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"genomics"
] |
2018
|
Initial acquisition and succession of the cystic fibrosis lung microbiome is associated with disease progression in infants and preschool children
|
Buruli ulcer ( BU ) is a skin infection caused by Mycobacterium ulcerans . The wounds of most BU patients are colonized with different microorganisms , including Staphylococcus aureus . This study investigated possible patient-to-patient transmission events of S . aureus during wound care in a health care center . S . aureus isolates from different BU patients with overlapping visits to the clinic were whole-genome sequenced and analyzed by a gene-by-gene approach using SeqSphere+ software . In addition , sequence data were screened for the presence of genes that conferred antibiotic resistance . SeqSphere+ analysis of whole-genome sequence data confirmed transmission of methicillin resistant S . aureus ( MRSA ) and methicillin susceptible S . aureus among patients that took place during wound care . Interestingly , our sequence data show that the investigated MRSA isolates carry a novel allele of the fexB gene conferring chloramphenicol resistance , which had thus far not been observed in S . aureus .
Buruli ulcer ( BU ) is a neglected necrotizing skin disease caused by Mycobacterium ulcerans , with the highest burden of the disease in West Africa , particularly in Benin , Cote d’Ivoire and Ghana [1] . The disease usually starts as a nodule , plaque , oedema or papule and progresses to form large ulcers with undermined edges if left untreated . It was previously shown that wounds of most BU patients are heavily colonized by many microorganisms , including Staphylococcus aureus [2 , 3] . S . aureus can be part of the human microbiota colonizing the skin and mucosal membranes without any clinical manifestations . However , once it crosses the skin barrier , or when the host immune system is compromised , this bacterium is able to cause a wide range of diseases , such as skin and soft tissue infections , osteomyelitis , pneumonia , meningitis , or bacteremia [4 , 5] . Therefore , S . aureus is considered a dangerous pathogen in both community-acquired and nosocomial infections . Colonization of healthy individuals with multi-drug resistant S . aureus is regarded as a risk factor for future development of S . aureus infections that are difficult to treat [6] . The S . aureus colonization of patients with a serious breach of skin barrier , such as patients with BU , burn wounds or the group of hereditary mechanobullous diseases epidermolysis bullosa ( EB ) , was previously shown to be very high [2 , 3 , 7–9] . Molecular typing of S . aureus isolated from the wounds of BU and EB patients has shown that their wounds often harbor multiple genotypes of this pathogen [3 , 10] . Recently , several S . aureus clones have been reported in health care institutions in Ghana with the sequence types ( ST ) 15 , 121 and 152 being the most prevalent as determined by multi-locus sequence typing ( MLST ) [11] . Notably , health care-associated infections ( HAIs ) caused by S . aureus impose a significant burden on patient care as a result of prolonged hospital stays , increased cost of treatments and high morbidity and mortality rates . Current practices implemented to reduce HAIs include cleaning of the hospital environment , hand hygiene and screening and decolonization of patients and health care workers [12–15] . Epidemiological data and molecular typing methods , such as pulsed-field gel electrophoresis ( PFGE ) , MLST , spa-typing , multiple-locus variable number tandem repeat fingerprinting ( MLVF ) , and whole-genome sequencing ( WGS ) of the infecting strains can be used to trace transmission events [16–19] . Each of these typing methods has particular advantages [20] . For example , MLVF is fast , cheap and highly discriminatory [21] , while WGS provides additional information on the genetic makeup of investigated isolates on top of a highly discriminatory typing result . BU patients may be at risk of hospital-associated colonization with S . aureus due to their frequent visits to particular health care centers for wound care . This represents an additional health risk for these patients , even if they are already colonized with community-acquired S . aureus . Therefore , the present study was aimed at uncovering possible S . aureus transmission events among BU patients using MLVF and WGS . Furthermore , WGS was applied to identify antimicrobial resistance ( AMR ) genes and to screen for mutations in genes that confer certain resistance phenotypes . The results obtained underpin the potential of the combined use of MLVF and WGS for the surveillance of S . aureus outbreaks in hospital settings .
The ethical committee of the Noguchi Memorial Institute for Medical Research ( NMIMR ) ( FEDERAL WIDE ASSURANCE FWA 00001824 ) approved the use of clinical samples for this investigation . Samples were collected upon written informed consent from adult subjects and a parent or guardian of any child participant on their behalf . A subset of the S . aureus isolates from BU patients that were previously collected and grouped by MLVF into thirteen clusters ( A-M ) [3] were selected for WGS . For the present study , isolates were selected from each of the thirteen MLVF clusters including two clusters suspected of patient-to-patient transmission events during wound care ( clusters H and F ) . Screening of BU patients for the presence of S . aureus had been repeated every two weeks for a period of seven months , which defined the sampling time points t1 to t13 in this study ( Table 1 ) . Patients involved in this screening were at different stages of the disease and treatment for BU . All presently investigated S . aureus isolates were obtained from positive anterior nares and wound cultures of eleven BU patients who attended the Pakro Health Center in the Eastern region of Ghana for antimicrobial therapy ( Table 1 ) . Genomic DNA was extracted from S . aureus isolates grown overnight on blood agar by using the Ultraclean microbial DNA isolation kit ( mo bio laboratories , Inc , Carlsbad , California , USA ) according to the manufacturers’ instructions . DNA libraries were prepared using the Nextera XT v2 kit ( Illumina , San Diego , CA , USA ) according to the manufacturers’ instructions and then run on a Miseq ( Illumina ) for generating paired-end 250-bp reads . De novo sequence assembly was performed using CLC Genomics Workbench v7 . 0 . 4 ( CLC bio A/S , Aarhus , Denmark ) after quality trimming ( Qs > 28 ) with optimal word sizes based on the maximum N50 value . The assembled files were imported as Fasta files into SeqSphere+ software version 1 . 1 ( Ridom GmbH ) . The sequence reads were submitted to the National Center for Biotechnology Information GenBank and are available under the BioProject PRJNA283747 and accession numbers: LGAE00000000 , LFTW00000000 , LFTV00000000 , LFTU00000000 , LFTT00000000 , LFOH00000000 , LFOG00000000 , LFNS00000000 , LFNR00000000 , LFNQ00000000 , LFNP00000000 , LFNO00000000 , LFNN00000000 , LFNM00000000 , LFNL00000000 , LFNK00000000 , LFNJ00000000 , LFNI00000000 , LFNH00000000 , LFMH00000000 , LFMG00000000 . The sequence data of the 21 isolates were characterized by using the core genome multilocus sequence typing ( cgMLST ) consisting of 1 , 861 genes and 706 S . aureus accessory genes . The complete sequence of each isolate was analyzed based on gene-by-gene comparison with the reference S . aureus strain COL ( GenBank accession no . NC_002951 ) and S . aureus cgMLST target definer function with the default parameters of the software as previously described [22] . Each allele was assigned a number and an allelic typing profile based on the combination of all alleles for each isolate by the software . A dendrogram of the sequenced isolates and two additional reference genomes that represent different sequence types ( ST5 [N315 GenBank accession no . BA000018 . 3] and ST8 [COL] ) was generated using an unweighted-pair group method using average linkages ( UPGMA ) . The concordance between the two typing methods was calculated with the Ridom EpiCompare software version 1 . 0 as described previously [19] . In this study a transmission event is defined to have occurred if the wound of a patient , previously not containing a particular S . aureus genotype , becomes colonized over time by an S . aureus with a genotype that is identical with the genotype of an S . aureus isolate collected from the wound of another patient . Here we investigated whether transmission events had indeed occurred during wound care of patients treated in the Pakro Health Center , using the SeqSphere+ scheme which assigned each S . aureus isolate an allelic typing profile as previously described . The typing profile will , subsequently , be known as cluster type ( CT ) . Hence a transmission event would have occurred if S . aureus isolates from different BU patients are grouped within the same CT . In our previous study , S . aureus isolates from BU out-patients , who visited the health center for wound care , were suspected to be involved in patient-to-patient transmission events [3] . These isolates were initially grouped by MLVF into clusters H and F ( Fig 1A ) . De novo assembled genome sequences of S . aureus isolates were queried against specific previously identified sequence features , or compared to complete S . aureus reference genomes with associated annotated genes ( S1 Table ) using blastN in the WebACT comparison tool with default settings ( http://www . webact . org/WebACT/prebuilt# ) . Further detailed analyses were performed with the Artemis Comparison Tool ( ACT ) software [23] . Specifically sequence data were queried for the presence of SCCmec elements and AMR genes . Similarity matches were filtered based on their length and percentage similarity scores , and only the filtered hits with at least 80% sequence similarity were then displayed by ACT and analyzed in detail . The AMR genes that were screened confer resistance to chloramphenicol , clindamycin , erythromycin , fusudic acid , kanamycin , lincosamide , methicillin , mupirocin , penicillin , rifampicin , streptogramin A and B , tetracycline , trimethoprim , tobramycin , and/or vancomycin . Antibiotic resistance profiles of the sequenced isolates were previously determined using vitek according to the EUCAST guidelines [3] .
From a total of 13 BU patients who visited the Pakro Healthcare Center for wound care 21 S . aureus isolates from the anterior nares ( n = 4 ) and wounds ( n = 17 ) were sequenced . These included six methicillin resistant MRSA and 15 methicillin susceptible S . aureus ( MSSA ) isolates . These isolates have been previously characterized by MLVF and spa-typing as shown in Table 2 [3] . A dendrogram was generated after SeqSphere+ analysis of the 21 sequenced isolates , which revealed 14 cluster types ( CTs ) ( Figs 1B and 2 ) denoted as 545 ( n = 4 ) , 714 ( n = 1 ) , 556 ( n = 1 ) , 546 ( n = 1 ) , 552 ( n = 1 ) , 554 ( n = 1 ) , 715 ( n = 1 ) , 549 ( n = 2 ) , 550 ( n = 1 ) , 553 ( n = 1 ) , 555 ( n = 1 ) , 547 ( n = 4 ) , 551 ( n = 1 ) and 548 ( n = 1 ) . This clustering by SeqSphere+ seemed to match well with the previous clustering of isolates by MLVF . To calculate the concordance between the SeqSphere+ and MLVF typing data , the Ridom epicompare software 1 . 0 ( Ridom GmbH ) was used with the Rands Adjusted co-efficient . This revealed a concordance of 0 . 924 . The 21 sequenced isolates were assigned to eight MLST types , namely ST1 , ST5 , ST15 , ST88 , ST121 , ST152 , ST508 , and the new ST3019 . The ST3019 is a single-locus variant of ST45 at the yqiL locus . A first transmission event was identified for four MRSA isolates belonging to ST88 , which were previously grouped in the MLVF cluster H ( Fig 1A ) [3] . These isolates were classified by SeqSphere+ as CT 545 ( Fig 1B ) . Within CT 545 the allelic profiles were identical ( Fig 2 ) . Of note , the four isolates were obtained from three different patients visiting the healthcare center over a period of seven months . The medical care for these patients involved antibiotic treatment and wound dressing changes ( Tables 1 and 2 ) . This particular MRSA was first identified in the wound of patient 2 , who tested negative at the first sampling time point ( t1 ) . Patient 2 was the first to start treatment in this study and was found to carry this particular S . aureus genotype at several sampling time points during treatment ( i . e . at t2 , t3 , t8 and t9 ) . Patients 7 and 19 started treatment 25 and 35 days later , respectively . They both visited the health care center for wound care at time point t2 . The wounds of patients 7 and 19 tested positive for S . aureus with the genotype of CT 545 at the sampling time points t5 ( patient 7 ) and t3 ( patient 19; Table 1 ) , which is indicative of transmission events . A second suspected transmission event was initially identified by MLVF typing ( cluster F ) and involved eight BU patients [3] . To investigate this possible transmission event in more detail four of the 25 isolates obtained from three patients were randomly selected and sequenced . These four MSSA ST152 isolates were assigned to CT 547 ( Fig 1B ) . The allelic profiles within this cluster differed by one ( Fig 2 ) . Patients 10 and 11 tested positive for S . aureus with this particular genotype at the same sampling time point ( t2 ) . Patient 10 remained positive for S . aureus with the CT 547 until sampling time point t8 , and patient 11 until time point t5 ( Table 1 ) . A third patient ( patient 3 ) was found to be positive for S . aureus with the CT 547 at sampling time point t8 ( Table 1 ) . Patients 5 , 6 , 7 , 18 and 24 became positive for S . aureus with this genotype at later time points than patients 10 and 11 . The patients 5 , 6 , 7 , 18 and 24 paid at least one visit at the health center for wound care that overlapped with visits by three other patients , which were found to be positive for the S . aureus genotype with the CT 547 ( Table 1 ) . It is noteworthy to mention that in each of the transmission events , the gene allele variation between isolates was not higher than one . This implies that the isolates were nearly identical with respect to their core genome . The assembled genomes of the 21 S . aureus isolates were used in blast comparisons to detect the presence of AMR genes , and the results are shown in Table 2 . Among the investigated isolates none was found to carry genes involved in resistance to erythromycin , fusidic acid , kanamycin , mupirocin , or vancomycin . Antibiotic resistance of the sequenced isolates was previously most often found against penicillin , chloramphenicol , tetracycline and trimethoprim [3] . Consistent with their penicillin resistance , all sequenced isolates carried various types of blaZ operons , which were located on chromosomally integrated transposons or plasmids . Specifically , the blaZ gene was found in 16 isolates that belonged to ST1 , ST5 , ST15 , ST88 , ST152 and ST3019 , while the blaZ-B variant was found in five isolates representing ST5 , ST508 and ST121 . Fourteen sequenced S . aureus isolates were chloramphenicol resistant of which six ( ST121 , ST3019 and ST152 ) carried various plasmids with a catA gene . Six other isolates ( ST88 and ST1 ) carried a novel allele of fexB that was not previously reported in S . aureus . In the case of one isolate , the phenotypic resistance for chloramphenicol could not be confirmed at the genomic level , which was potentially due to the loss of the resistance gene . Resistance to rifampicin was identified in one isolate belonging to ST121 where the rpoB gene was found to encode an amino acid substitution that changed Asp471 into Gly . Resistance to tetracycline was identified in 16 isolates , which was confirmed by the identification of resistance genes , such as tetK , tetL and tetM . The tetK gene was located on plasmid pT181 , which was found in 10 isolates representing different STs . Five isolates of ST88 contained the tetL and tetM genes located on identical mobile genetic elements integrated into their genomes , while one isolate of ST15 contained a transposon with tetM . The presence of a plasmid or transposon carrying the drfG gene responsible for trimethoprim resistance was detected in five isolates that belonged to ST5 , ST15 and ST121 . Resistance to streptomycin was limited to three isolates of ST152 where the str gene was present . Of the six methicillin resistant isolates , five belonging to ST88 contained the mecA gene , whereas one ST5 isolate contained neither mecA nor mecC . The latter isolate was termed borderline oxacillin resistant S . aureus ( BORSA ) . Intriguingly , the BORSA isolate contained no mutations in the genes for the penicillin-binding proteins PBP1 , PBP2 , and PBP3 or the YjbH protein , which were previously proposed to be involved in BORSA phenotypes [24] . However , sequence comparisons revealed that the PBP2 protein of the BORSA isolate contains a Tyr residue at position 197 , while the PBP2 protein of S . aureus N315 contains a Cys residue at this position . Furthermore , the BORSA isolate showed resistance to fluoroquinolones , which may be due to a specific mutation in the gyrase A gene ( Ser84Leu ) .
In the present study , we have investigated S . aureus transmission events in BU patients during wound care by implementing a WGS-based gene-by-gene typing approach using SeqSphere+ . The SeqSphere+ scheme grouped the 21 sequenced S . aureus isolates into eight different STs . Sequenced S . aureus that belonged to ST88 isolates shared identical characteristics ( spa-types t186/t786 , SCCmec type IVa , PVL-negative ) with isolates collected from out-patients in Egypt and Angola , indicating a larger geographic distribution on the African continent [25 , 26] . The new ST3019 ( spa-type t939 ) identified in this study belongs to the same clonal complex ( CC45 ) as ST45 and ST508 . Compared to ST45 a single locus variation was observed at the yqiL locus for ST3019 and in the aroE locus for ST508 . Using SeqSphere+ , we identified two major clusters of S . aureus isolates from different BU patients , which may reflect transmission events that occurred during overlapping visits to the Pakro Healthcare Center where these patients received wound care . None of these patients carried S . aureus with the CTs 545 or 547 on their first visit to the Pakro Healthcare Center , strongly suggesting that they acquired the respective S . aureus types upon wound care . Interestingly , the majority of S . aureus isolates from BU patients belong to lineages characterized by spa-types t786 and t355 that have been already reported in health care settings in Ghana [11] . This suggests the nosocomial acquisition of these S . aureus types by patient-to-patient transmission between BU patients and healthcare workers that may have occurred due to inadequate hygiene . Indeed , it has been reported in a recent study that 8 of 11 MRSA transmission events among patients in intensive care settings were potentially due to poor hand hygiene [16] . This could probably be avoided by wearing gloves and protective gowns , and strict implementation of hand hygiene [27–29] . Basic preventive measures , such as adherence to aseptic techniques may further reduce the risk of infection thereby improving wound care of patients , provided that gloves , gowns , adequate dressing materials , running water and hand rub alcohol are made available . With a steady supply and stock of equipment and disposables , routine screening of patients and healthcare workers for S . aureus may be less critical . Genotypic data of the isolates sequenced confirmed the results of the antimicrobial resistance profiles described previously [3] . Interestingly , the chloramphenicol resistance of some isolates was conveyed by the fexB gene ( Table 2 ) , which was thus far not encountered in S . aureus . On the other hand , fexB was previously reported in Enterococcus faecium EFM-1 and Enterococcus hirae EH-1 isolates from pigs [30] . As Enterococci were previously identified in the wounds of BU patients , it is conceivable that the MRSA isolates acquired the fexB gene by horizontal gene transfer from such Enterococci [3] . Furthermore , a BORSA phenotype was identified in an isolate belonging to ST5 . Such a BORSA phenotype was previously reported for S . aureus isolates with ST1 , ST8 and ST15 that were implicated in wound infections in Scotland [24] . The presence of specific mutations in the genes coding for four proteins , namely PBP1 , PBP2 , PBP3 and YjbH , was proposed to be involved in the BORSA phenotype . However , in the genome sequence of the presently investigated BORSA isolate from a BU patient , none of these mutations was found . After genomic comparison of the BORSA isolate with the N315 reference genome , the only difference was observed for PBP2 , where at position 197 a cysteine residue was replaced by a tyrosine residue . However , this PBP2 amino acid substitution is encoded by the majority of S . aureus genomes available in the NCBI database and , therefore , it may not explain the BORSA phenotype observed . Further comparative genome analyses revealed about 300 additional non-synonymous SNPs , which could contribute to the observed BORSA phenotype . In summary , WGS of S . aureus isolates from BU patients and the subsequent analysis of sequencing data using the SeqSphere+ scheme revealed likely patient-to-patient transmission events in a healthcare setting in Ghana . This indicates a need for the implementation of improved hygiene protocols in healthcare settings where BU patients receive wound care . Apart from the detection of transmission events , WGS has the advantage that it also provides information on antimicrobial resistance . Related to the antimicrobial resistance pheno- and genotypes identified in S . aureus isolates from BU patients , it is important to bear in mind that antimicrobial pressure has the potential to aggravate resistance , with an inherent risk for transmission of resistant organisms . Therefore , even in low-resource settings , antimicrobial stewardship programs are likely to have added value , with more restrictive antimicrobial use than currently practiced [2] .
|
Buruli ulcer ( BU ) is a skin infection caused by Mycobacterium ulcerans . The wounds of most BU patients are colonized with different microorganisms , including Staphylococcus aureus . This study investigated patient-to-patient transmission events during wound care in a health care center . S . aureus isolates from patients who visited the health center at the same time points were analyzed using whole-genome sequencing . Analysis of sequence data confirmed transmission of methicillin resistant S . aureus and methicillin susceptible S . aureus among patients that took place during wound care .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Molecular Characterization of Staphylococcus aureus Isolates Transmitted between Patients with Buruli Ulcer
|
The most severe clinical form of neurocysticercosis ( NC ) occurs when cysticerci are located in the subarachnoid space at the base of the brain ( SaB ) . The diagnosis , monitoring and treatment of NC-SaB , constitutes a severe clinical challenge . Herein we evaluate the potential of the HP10 antigen detection enzyme-linked immunosorbent assay ( HP10 Ag-ELISA ) in the long term follow-up of NC-SaB cases . Assay performance was compared with that of Magnetic Resonance Imaging ( MRI ) . In addition , the robustness of the HP10 Ag-ELISA was evaluated independently at two different institutions . A double-blind prospective cohort trial was conducted involving 38 NC-SaB cases and a total of 108 paired serum and cerebrospinal fluid ( CSF ) samples taken at intervals of 4 to 8 months for up to 43 months . At each medical visit , results of sera and CSF HP10 Ag-ELISA and MRI obtained at last visit were compared and their accuracy was evaluated retrospectively , considering radiological evolution between appointments . In the long-term follow-up study , HP10 Ag-ELISA had a better agreement than MRI with retrospective radiological evaluation . High reproducibility of HP10 Ag-ELISA between laboratories was also demonstrated . Results reported in this study establish for the first time the usefulness of the comparatively low cost HP10 Ag-ELISA for long term follow-up of NC-SaB patients .
Neurocysticercosis ( NC ) is one of the most frequent parasitic diseases affecting the human central nervous system [1] . It is transmitted by the ingestion of Taenia solium eggs mainly through the consumption of contaminated vegetables , and is still prevalent in most countries of Asia , Africa and Latin America , including México [2] . Additionally , its prevalence is rising in the United States and some European countries due to increasing immigration [3] , [4] . NC severity critically depends on the location of the parasite . The disease is mostly benign when cysticerci are located in the cerebral parenchyma , neuroimaging techniques accurately indicating the number , localization , viability of the parasites and the intensity of the inflammatory reaction [5] , [6] . In contrast , when parasites are located in the basal subarachnoid space ( NC-SaB ) , clinical presentation is generally severe , cysticidal drugs are less effective and neuroradiological studies are less precise , diagnosis relying mostly on indirect clues such as the enlargement of the basal cisternae [7] , [8] . Furthermore , neuroradiological studies represent the most expensive healthcare-related costs [9] , and are only available at major urban centres whereas the principal population at risk is mostly rural . Detection of the secreted metacestode antigens , particularly HP10 [10] is becoming an increasingly accepted test for diagnosing severe NC [11]–[19] . Previous studies have demonstrated the high specificity and sensitivity of HP10 antigen ELISA assay to detect NC-SaB [14]–[18] , and its similar accuracy either when sera or CSF samples are employed [12] , [15] , [18] . In this first prospective long-term study focusing on NC-SaB , we evaluated the assay reproducibility and accuracy , comparing the MRI and HP10 results and considering the radiological evolution of the patients retrospectively at each medical appointment .
This prospective longitudinal study was performed in a total of thirty-eight NC-SaB patients who attended to the Instituto Nacional de Neurología y Neurocirugía ( INNN ) , Mexico City . The patients included were selected consecutively between August 2008 and March 2010 . Initially , fifty patients were included , but 8 were lost in the follow-up as they did not come to the second medical appointment and , in 4 cases , paired CSF and serum HP10 determinations were not made as their increased intracranial pressure precluded the taking of CSF samples . Diagnosis was based on clinical manifestations ( presence of focal deficit , affection of cranial nerves and intracranial hypertension ) , imaging studies ( MRI with images compatible with the presence of cysticerci , ie . mainly , enlargement or deformation of a basal cistern or visualization of cystic vesicles ) , and HP10 positive in CSF . Patients were followed-up during 6 to 43 months , resulting in a total of 108 individual clinical , radiological , and serological evaluations . Each of the 38 cases had received between 1 to 7 cycles of cysticidal drug treatment before being included in the trial . In most cases albendazole ( ABZ ) was used at a dose of 30 mg/kg/day during 8 consecutive days . Few patients received praziquantel ( PZQ , 50 mg/kg/day during 10 days ) or a combination of ABZ plus PZQ . Cysticidal drugs were always accompanied with prednisone ( 1 mg/kg/day ) or dexamethasone ( 0 . 4 mg/kg/day ) , followed by individual decreasing prednisone doses , depending on the clinical and inflammatory status . MRI results ( gadolinium-enhanced T1 and T2- weighted and FLAIR , ( Fluid Attenuation Inversion Recovery ) , MRI sequences ) were recorded , and paired CSF/serum samples were tested for HP10 Ag-ELISA . Follow-up was carried out at least once in all cases , at 4–8 month intervals , with further combined HP10 Ag-ELISA and MRI studies . The time between CSF/serum sampling for HP10 Ag-ELISA and the associated MRI examination was ≤1 month . During this time , no cestocidal treatments were administered . At each medical visit , changes in MRI studies since last appointment were assessed and , based on these criteria , the accuracy of MRI and HP10 results at last evaluation was established . We considered that patient were free of SaB vesicular parasites at last evaluation if we did not see any changes in the suspect images on MRI ( absence of disappearance or diminution in case of specific treatment or enlargement if no cestocidal treatment was administered ) . On the other hand , we concluded that patients present SaB vesicular parasites at their last evaluation if we observed a reduction of the suspected image after treatment or its enlargement if treatment had not been administered . All MRI studies were double-blind interpreted by a certified neuroradiologist with extensive experience in NC diagnosis . MRI was considered positive if a vesicular cyst located in the subarachnoid basal cisterns or a deformation of a cistern compatible with the presence of a vesicular cyst were observed . The study was approved by and carried out under the guidelines of the Ethical Committee of the Instituto Nacional de Neurología y Neurocirugía , D . F . , México . All patients provided written informed consent for the collection of samples and subsequent analysis . HP10 Ag-ELISA were blindly performed by two Mexican laboratories ( Instituto de Investigaciones Biomedicas , UNAM ( IIB ) , and INNN ) , both in Mexico City , Mexico . Afterwards , sample codes were revealed and results were analyzed . HP10 antigen was detected by Ag-ELISA as described previously [10] . All samples were run in duplicate by two experimented technicians . Briefly , plates ( Nunc , Rochester , New York , USA ) were coated with monoclonal antibody ( MoAb ) HP10 ( 100 µl at 10 µg/ml in 0 . 07 M NaCl buffered with 0 . 1 M borate , pH 8 . 2 ) and left overnight at 4°C , washed four times with 200 µl/well of wash solution ( 0 . 9% w/v NaCl containing 0 . 05%v/v Tween 20 ) and then blocked using 200 µl of phosphate-buffered saline containing bovine serum albumin ( Roche , México ) ( 1 . 0% w/v and 0 . 05% v/v Tween 20 ) and left by 60 min at room temperature before being washed in a similar way . Undiluted CSF or serum samples ( 100 µl/well ) were added and incubated 30 min at 37°C . Bound HP10 parasite antigen was detected using biotinylated MoAb HP10 ( 2 µg/ml in diluent , for 60 min at 37°C ) , horseradish peroxidase-conjugated streptavidin ( Zymed , San Francisco , California , USA ) ( 1∶4000 in diluent , 30 min at 37°C ) and tetramethylbenzidine ( Zymed , San Francisco , California , USA ) as substrate . Color reaction was allowed to proceed for 30 min at 4°C in the dark and was stopped by adding 100 µl 0 . 2 M H2SO4 ( Baker , Estado de Mexico , Mexico ) . Optical densities ( OD; 450 nm ) were determined in an ELISA processor ( at INNN: Bio-Rad Microplate Reader Benchmark , Hercules , California , USA; at IIBM: Opsys MR Dynex Technology , Chantilly , Virginia , USA ) . A sample was considered positive if the mean OD at 450 nm value was higher than the cut off value , which was calculated based on the mean of the OD plus 2 SD of CSF and sera from non-NC controls . A database was built using Excel 7 . 0 ( Microsoft ) software . MRI findings and CSF/serum HP10 levels were recorded . Statistical analysis was made using SPSS 10 . 0 ( Microsoft ) , Epidat 3 . 0 and SAS 9 . 0 softwares . Parametric statistics ( mean and SD ) were calculated . Kappa coefficient ( k ) and 95% confidence intervals were calculated to evaluate the qualitative agreement between HP10 results in CSF and sera and between institutions . This coefficient varied between −1 and +1 . The closer the value is to1 , the stronger the agreement [20] . Generalized Estimating Equations ( GEE ) analysis [21] was performed in order to confirm agreement taking into account repeated observations and variability of period of time between appointments . Results confirm that these two factors did not modify the agreement between tests .
The main features of the 38 patients included in the study are summarized in table 1 . Most patients had vesicular cysticerci located in the SaB with inflammatory CSF , in spite of the previous cysticidal treatment . Kappa analysis indicated a good to very good level of agreement between positive/negative allocations comparing results obtained in both institutions ( CSF Kappa: 0 . 86 [0 . 75–0 . 97 , P<0 . 0001] and serum Kappa: 0 . 76 [0 . 64–0 . 89 , P<0 . 0001] ) [15] . A good level of agreement15 between positive/negative allocations for paired CSF and serum samples ( CSF/serum IIBM Kappa: 0 . 63 [0 . 48–0 . 79 , P<0 . 0001] and CSF/serum INNN Kappa: 0 . 64 [0 . 49–0 . 79 , P<0 . 0001] ) was found . There was complete agreement between radiological and CSF HP10 Ag-ELISA results in 18 out of the 38 cases ( 50 samples ) . As shown in table 2 , as judged by negative MRI and CSF HP10 Ag-ELISA results , the infection resolved only in three of these cases . In the other 15 cases , cysticerci persisted and so additional ABZ cycles were indicated . Of the remaining 20 patients without complete agreement between MRI and CSF HP10 Ag-ELISA assays in all samples , discordant results were found in 32 of the 58 samples ( 55 . 2% ) . As shown in table 3 , a complete agreement between HP10 Ag-ELISA in sera and MRI evaluation was obtained in 14 patients , corresponding to 38 paired samples . In the remaining 24 patients , disagreement was observed in some of the samples ( 43 of 70 , 61% ) . As shown in table 4 , agreement between MRI results and retrospective evaluation taking in count radiological evolution between appointment was moderate: Kappa: 0 . 45 ( 0 . 26–0 . 63 ) . In 24 samples , divergent results occurred: 15 of them with positive results in MRI and negative results in retrospective evaluation , and the reverse in 9 of them . As shown in table 5 , agreement between CSF HP10 Ag-ELISA results and retrospective evaluation taking in count clinical and radiological evolution was very good: Kappa: 0 . 82 ( 0 . 70–0 . 94 ) . Differences between the evaluations occurred in only 8 samples , with positive results in HP10 Ag-ELISA/negative results in retrospective clinical/radiological evaluations in 6 and the reverse in two . As shown in table 6 , the agreement between sera HP10 Ag-ELISA results and retrospective evaluation taking in count clinical and radiological evaluation was moderate: Kappa: 0 . 56 ( 0 . 39–0 . 73 ) . Differences between the evaluations occurred in 21 samples , with positive results in sera HP10 Ag-ELISA/negative results in retrospective clinical/radiological evaluations in 8 and the reverse in 13 .
Diagnosis of NC- SaB is still a challenge . In previous studies , it was demonstrated that the levels of secreted cysticercal HP10 antigen in CSF and serum is an accurate method to diagnose vesicular cysticerci located in the ventricles or SaB [11]–[15] . HP10 Ag-ELISA was originally designed to diagnose Taenia saginata cysticercosis [10] , but since HP10 antigen is shared by other cestodes such as T . solium , it is also useful for NC diagnosis [11]–[15] . In this prospective longitudinal study , we evaluated the usefulness of the HP10 Ag-ELISA assay in the follow-up of patients with NC-SaB , and compared its predictive capacity with MRI , the best tool currently available for the diagnosis and follow-up of these patients . Both tests were compared with a retrospective diagnosis based on radiological evolution between medical appointments . The results obtained herein demonstrate the usefulness of HP10 Ag-ELISA for the follow-up of severe NC-SaB patients . Of the 108 paired MRI/HP10 Ag-ELISA evaluations , agreement between MRI and CSF HP10 Ag-ELISA was found in 76 ( 70 . 4% ) and between MRI and sera HP10 in 65 ( 60% ) . Retrospective analysis evaluating at each medical appointment the radiological evolution between visits shows that this lack of agreement appears to be principally due to a misinterpreted MRI rather than to serum and/or CSF HP10 Ag-ELISA assay . In addition , HP10 Ag-ELISA proved to be a highly reproducible method , as high Kappa coefficients of 0 . 76 and 0 . 86 in sera and CSF , respectively , were obtained when comparing results obtained in two different Mexican institutions . Another result meriting comment is the confirmation that determination of HP10 in sera is of interest , giving quite similar agreement than MRI with retrospective diagnosis . Agreement between CSF HP10 determination and retrospective evaluation was higher , but considering the invasive nature of the lumbar puncture , procedure necessary to collect CSF , the use of sera instead of CSF could be recommended , particularly if there is no clinical emergency , and also for monitoring of SaB neurocysticercosis patients from rural communities where imaging facilities are not available . Considering these results , it is possible to propose HP10 determination in order to reduce the number of MRI required in the follow-up of such patients . Generally , MRIs are realized each 6 months to evaluate response to treatment , and frequently 4 to more than 8 studies are made by patient during their illness and treatment . The results herein presented show that it will be possible to reduce the number of studies to at least a half without taking a risk of misdiagnosis . We recommend that after diagnosis of NC-SaB by MRI ( diagnosis must be made by MRI as it will permit the evaluation of the extent of the disease ) , the follow-up of the patients could be made by one MRI and one HP10 evaluation each year until parasites disappears . Election of serum or CSF HP10 evaluation , as said before , will depend on the gravity of the patients and the possibility to make lumbar puncture without risk . Importantly , the results obtained in this study point to the need for improving MRI for NC diagnosis . As shown here , when cysticerci are located in the SaB , diagnostic radiology techniques are imprecise . This can be explained by the fact that parasites emit a signal of similar intensity than the CSF itself . In addition , in most cases the image is not enhanced by administration of intravenous contrast , and finally , metacestodes commonly lack the distinguishing scolex that allows their identification [8] . In this respect , new MRI technologies based on a fast imaging employing steady-state acquisition ( FIESTA ) has shown to have a good capacity to diagnose intraventricular cysts because of their high spatial resolution and signal-to-noise rate [22] . Nevertheless , despite the promising results reported , these procedures are not yet standardized for NC diagnosis . It is also interesting to note that FLAIR sequences , permitting better visualization of the scolex and the cyst wall , did not resolve all the cases in our studies [5] , [23] . The need for an economic and reliable diagnosis of cysticercosis is now of urgent concern in rural cysticercosis endemic communities , particularly as recent plans for the control of neglected tropical diseases , such as schistosomiasis , involve the mass and indiscriminate dosing of entire African populations with antihelmintic drugs , including praziquantel [24] . Such treatment , of population co-infected with T . solium neurocysticercosis , would be predicted to increase the inflammation around the cyst , thereby provoking severe side effects [25]–[27] . Thus a prior screening for cysticercosis/neurocysticercosis , ideally including antibody and antigen detection , might be advisable . Neuroradiological studies required for diagnosis and patient follow-up currently represent the main cost of NC management in particular in severe NC-SaB , which usually requires multiple cysticidal cycles of treatment [9] . Thus , the alternative use of a serological non-invasive , non-expensive , highly accurate assay could have a real economic impact . To illustrate this point , is worthy to mention that an MRI in Mexico costs approximately 400 dollars , a cost dramatically contrasting with approximately 1–10 dollars for a typical commercial Ag-ELISA test . HP10 ELISA could have an undeniably positive economic impact for the patient and for the health institutions as well . This study offers arguments to strongly recommend its routine use for the follow-up of these severely affected patients . Despite the HP10 Ag-ELISA accuracy , HP10 false positive results would lead to unnecessary treatment of these patients ( being corticoid administration the main problem ) . Thus , it is urgent to investigate the factors underlying false positive results . False negative HP10 results were observed only in 2 of the 108 evaluations ( 1 . 8% ) in CSF and in 13 ( 12% ) in serum , posing a very low risk of failing to discriminate a NC-SaB patient still requiring treatment . To avoid the risk of incomplete drug treatment , we suggest repeating serum HP10 Ag-ELISA test in patients presenting NC-SaB with HP10-negative results after treatment . It is important to note that these results must be confirmed . Limitations of this study are mainly due to: 1 ) the small number of patients that leads to large confidence intervals and uncertainty . However , it must be stated that these types of patients ( NC-SaB ) are infrequent and that this study is the first one with such large follow-up; 2 ) the retrospective diagnosis can be criticized as it is known that evolution of parasites is very slow . It is possible that in some cases , for example , radiological picture of the patients did not change between 2 appointments mistakenly making believe that there were no parasites although there were present . We are conscious of this fact and we hope that the standardization of new MRI techniques will permit to give new tools to evaluate these tests . In conclusion , this study establishes the usefulness and economic advantage of the HP10 Ag-ELISA applied on CSF and serum samples for the follow-up of patients with NC-SaB , the most severe form of the disease . Hopefully , these results will lead to the rapid commercialization of a HP10 antigen diagnostic kit to favor its employment worldwide .
|
Neurocysticercosis is one of the most frequent parasitic diseases affecting the human central nervous system . The most severe clinical forms occur when parasites are located in the subarachnoid space at the base of the brain . In these instances , cysticidal drug efficacy is reduced and neuroimaging studies are less reliable as diagnostic tools . Previous works highlighted the value of antigen detection by ELISA test to detect viable parasites in these locations . In this prospective study , we evaluate its utility in patient follow-up , comparing its performance with magnetic resonance imaging results . Results from both procedures were also compared retrospectively at each medical appointment considering radiological evolution since last evaluation . Thirty-eight patients were included , with a total of 108 samples collected over 43 months . We demonstrated that antigen detection in these patients is an accurate tool in determining the efficacy of cysticidal treatment . This result is of great potential , considering the difficulty for the patients in endemic countries to access imaging studies and the much lower cost of the assay with respect to magnetic resonance imaging .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"infectious",
"diseases",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"clinical",
"laboratory",
"sciences",
"diagnostic",
"medicine",
"clinical",
"immunology",
"neurological",
"disorders",
"cysticercosis",
"neurology",
"neglected",
"tropical",
"diseases",
"immunology",
"parasitic",
"diseases"
] |
2013
|
Neurocysticercosis: HP10 Antigen Detection Is Useful for the Follow-up of the Severe Patients
|
Blastocystis is an extracellular , enteric pathogen that induces intestinal disorders in a range of hosts including humans . Recent studies have identified potential parasite virulence factors in and host responses to this parasite; however , little is known about Blastocystis-host attachment , which is crucial for colonization and virulence of luminal stages . By utilizing 7 different strains of the parasite belonging to two clinically relevant subtypes ST-4 and ST-7 , we investigated Blastocystis-enterocyte adhesion and its association with parasite-induced epithelial barrier disruption . We also suggest that drug resistance in ST-7 strains might result in fitness cost that manifested as impairment of parasite adhesion and , consequently , virulence . ST-7 parasites were generally highly adhesive to Caco-2 cells and preferred binding to intercellular junctions . These strains also induced disruption of ZO-1 and occludin tight junction proteins as well as increased dextran-FITC flux across epithelial monolayers . Interestingly , their adhesion was correlated with metronidazole ( Mz ) susceptibility . Mz resistant ( Mzr ) strains were found to be less pathogenic , owing to compromised adhesion . Moreover , tolerance of nitrosative stress was also reduced in the Mzr strains . In conclusion , the findings indicate that Blastocystis attaches to intestinal epithelium and leads to epithelial barrier dysfunction and that drug resistance might entail a fitness cost in parasite virulence by limiting entero-adhesiveness . This is the first study of the cellular basis for strain-to-strain variation in parasite pathogenicity . Intra- and inter-subtype variability in cytopathogenicity provides a possible explanation for the diverse clinical outcomes of Blastocystis infections .
Blastocystis is a unicellular , genetically diverse protist , phylogenetically placed within the Stramenopiles [1] , and it is the only member of this group associated with human pathological changes [2] , [3] . It is a species complex comprising 17 subtypes ( STs ) at least 9 of which are found in humans [4] , [5] . The prevalence of this parasite is usually higher in developing countries , ranging from 30% to 50% , and 1 . 5% to 15% in developed countries [2] , [6] . However , some select populations in developed countries may have much higher prevalence [7] . It is frequently reported in human fecal samples from symptomatic patients as well as healthy individuals [8]–[10] . The parasite induces enteritis , manifested in mucous and watery diarrhea , bloating , abdominal pain and/or vomiting [3] . Clinical studies also associate Blastocystis with other intestinal and dermatological inflammatory disorders , such as irritable bowel syndrome and urticaria [11]–[14] , respectively . Patients immunocompromised due to HIV or cancer are particularly susceptible to infections [15]–[18] , suggesting that Blastocystis is also an opportunistic pathogen . Despite being discovered more than 100 years ago [19]–[21] , it is difficult to argue the clinical significance and pathogenic potential of Blastocystis [21] , since infections do not consistently lead to intestinal symptoms [22]–[24] . While a large number of infected individuals present with clinical symptoms [22] , [25] , asymptomatic carriage of the parasite is also common [26] . Moreover , in symptomatic patients , the duration and severity of symptoms vary from acute enteritis to chronic , mild diarrhea [24] , [27] . There is no consensus on the reasons for the observed diverse intestinal symptoms . A number of reports have suggested a strain- or subtype-dependent variation in parasite pathogenicity [1] , [24] , [27]–[31] . Studies associate ST-1 , -4 , and -7 with pathological alterations in humans , whereas ST-2 and ST-3 are considered apathogenic [22] , [24] , [28] . Also , the presence of both pathogenic and apathogenic strains within one subtype has been reported [32] , [33] . Yet , the factors determining this variation in pathogenicity across different Blastocystis strains have not been resolved . Although recent advances have been made in the knowledge of its molecular and cellular biology [3] , [21] , [34] , [35] , as well as pathogenic mechanisms [30] , [36] , many gaps remain unfilled regarding the pathogenesis of Blastocystis . The chain of events leading to parasite-induced pathological changes are largely unknown , the most important early event being Blastocystis-enterocyte contact [37] . Extracellular , enteric pathogens have to overcome the intestinal luminal peristaltic flow to remain in the intestine [38] , [39] . Adhesion to epithelial cell surfaces is thus a critical step for their infection and effective colonization of the host [40] . Studies have suggested the level of adhesion is directly linked to the virulence properties of pathogens [40] , [41] . Highly adhesive strains of Giardia , Entamoeba , Trichomonas and other eukaryotes have been shown to result in more severe damage of the epithelium compared with less adherent strains [42]–[44] . Unlike other luminal parasites , Blastocystis is immotile [26]; hence , efficient anchoring to epithelial cells is even more crucial for its survival in the host gut as well as the induction of entero-pathogenesis . The ability of Blastocystis to adhere to the intestinal epithelium has not yet been investigated . Clearly , it is important to determine adhesiveness of the parasite with enterocytes across different Blastocystis strains and investigate its association with parasite pathogenicity . Another issue complicating the pathogenic potential of Blastocystis is reports of treatment failure [25] , [27] , [45]–[48] . Although metronidazole is the treatment of choice , physicians are often skeptical about prescribing antibiotics for Blastocystis infections due to frequent reports of non-responsiveness to chemotherapy [24] . Strain-to-strain variation within Blastocystis in susceptibility to Mz and other antiparasitic agents among Blastocystis strains is commonly reported [49] , [50] , and has been proposed to be the reason for frequent treatment failures in parasite infections [25] , [49] . However , from an evolutionary standpoint , mutations associated with drug resistance may impair essential biological functions or impose energy demands on the organism , leading to decreased microbial fitness [51] , [52] . Studies of a variety of pathogens , including different species of viruses , bacteria and parasites , indicate that antimicrobial resistance places a toll on the organisms' fitness as well as virulence [53] . A recent study in an intestinal protozoan parasite , Giardia , revealed impaired attachment and decreased infectivity in Mz resistant ( Mzr ) strains compared with parental Mz sensitive ( Mzs ) strains [54] and was given as a possible reason for the scarcity of treatment failure in people with symptomatic parasite infections . The interplay among drug resistance , fitness and virulence in Blastocystis has never been studied . Considering the frequent reports of treatment failure in humans with symptomatic Blastocystis infections , it will be interesting to establish whether drug resistance exerts any effects on the pathogenicity as well as other aspects of parasite fitness . In the present study , using seven isolates from two clinically important subtypes Blastocystis ST-4 and ST-7 , we first demonstrated extensive intra- and inter-subtype variability in inducing intestinal barrier dysfunction , and then investigated whether adhesion contributed to this variation . We found that Mz resistance correlates with impairment of parasite adhesion and adhesion-associated cytopathic effects . Additionally , we also showed that Mz resistance was associated with nitrosative stress tolerance , an important factor assisting parasite survival in the gut lumen .
All Blastocystis-host interaction experiments were performed using Caco-2 human colonic cell line ( ATCC ) . Caco-2 stock cultures were maintained in T-75 flasks in a humidified incubator with 5% CO2 at 37°C . Cell cultures were grown in Dulbecco's modified Eagle's medium ( HyClone ) supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) and 1% each of sodium pyruvate and MEM , antibiotic Penicillin-Streptomycin ( Gibco; final concentrations are 1 , 000 units/mL of penicillin and 1 , 000 µg/mL of streptomycin ) . Culture health was evaluated using trypan blue assay; only cultures with >95% viability were used for the experiments . Cells were trypsinzed with 0 . 25% trypsin-EDTA for subculture and cell seeding . For Western blot analysis , Caco-2 monolayers were grown on 6-well cell culture plates ( Corning ) until 100% confluency . For immunofluorescence and confocal microscopy , Caco-2 monolayers were cultured on Poly-L-lysine-treated 15 mm glass coverslips placed in standard 24-well culture plates . For epithelial permeability experiments , cells were grown on Millipore transwell filters with PET membranes of a 3 µm pore-size , placed in 24-well tissue culture plates . In order to synchronize cells before experiments , all cultures were serum-starved overnight in antibiotic- and serum-free DMEM . Seven axenized isolates of Blastocystis ( designated B , C , E , G , H , S-1 and WR-1 ) were used in the present study ( Figure S1 ) . All seven isolates were subtyped previously by small-subunit of ribosomal RNA gene analyses [55] . Isolates B , C , E , G and H were originally recovered from symptomatic patients at the Singapore General Hospital [56] and they all belonged to ST-7 according to recent classification system [57] . Isolates S-1 and WR-1 were isolated from a rat during an animal survey [58] and they belonged to ST-4 . Both ST-4 and ST-7 are well characterized zoonotic Blastocystis isolates commonly detected in humans with gastrointestinal symptoms [24] . Stock cultures of all seven isolates were maintained under the same conditions as described previously [49] . In brief , the parasites were maintained in 10 ml of pre-reduced Iscove's modified Dulbecco's medium ( IMDM ) containing 10% heat-inactivated horse serum in an anaerobic jar ( Oxoid ) with an AnaeroGen gas pack ( Oxoid ) at 37°C . All Blastocystis isolates used in this study were obtained from an existing collection at the Department of Microbiology of the National University of Singapore ( NUS ) . Human isolates were obtained from patients at the Singapore General Hospital in the early 1990s , before Institutional Review Board was established in NUS . All samples were anonymized . The optimized Blastocystis viability assay was used to measure 50% inhibitory concentrations ( IC50s ) of different drugs for the Blastocystis strains [49] . Briefly , stock solutions of drugs were prepared in dimethyl sulfoxide ( DMSO ) , diluted in pre-reduced Blastocystis culture medium , transferred to 96-well plates and pre-reduced in anaerobic jar for 4 h at 37°C . Parasites were counted and 0 . 5×106 cells were incubated in each well of a standard 96-well plate with dilutions of different drugs ranging between 0 and 100 µg/ml . The drug concentration range was reduced or increased depending on the results of the first test . A range of parasite counts between zero and 0 . 5×106 per well were used for viability control . The final DMSO concentration was kept constant at 0 . 5% in each well . Since the parasite redox activity varies with volume , the final total volume of each well was kept constant at 200 µl . After 24 h of drug exposure , resazurin solution ( Sigma ) was added to each well at a final concentration of 10% ( v/v ) ; 3 h after incubation , under anaerobic conditions at 37°C , fluorescence readings of resazurin were taken at 550-nm excitation and 580-nm emission wave lengths using a Tecan Infinite M200 reader . Values were imported into GraphPad Prism5 software and IC50s were calculated . The growth curve of each of the seven isolates was made by culturing over a period of 96 h . Briefly , 5×106 parasites of each isolate were inoculated into IMDM and cultured under anaerobic conditions at 37°C . At each of the five time points ( 0 , 24 , 48 , 72 and 96 h ) , parasite pellets were resuspended and cell numbers were counted using appropriate dilutions in a hemocytometer . Parasite protease activity was determined using an azocasein assay [59] . Briefly , lysates of 4×106 parasites were co-incubated with 2 mM dithiothreitol ( DTT ) ( Sigma ) at 37°C for 10 min to activate protease activity . 100 µl of a 5 mg/ml solution of azocasein ( Sigma ) was prepared in PBS ( pH 7 . 4 ) and incubated with 100 µl of parasite lysate for 1 h at 37°C . The reaction was stopped by adding 300 µl of 10% trichloroacetic acid ( TCA ) , and samples were incubated on ice for 30 min . Undigested azocasein was removed by centrifugation ( 5 , 000×g for 5 min ) , and the resultant supernatant was transferred to a clean tube containing 500 µl of 525 mM NaOH . Absorbance was measured using a spectrophotometer at 442 nm ( Tecan Magellan ) . PBS was used as a negative control . For inhibition experiment , 2 mM iodoacetamide ( IA ) were added to the parasite lysate and incubated for 1 h at room temperature ( 22–24°C ) to inhibit the cysteine proteases [60] . One-day-old Blastocystis cells were collected and stained with carboxyfluorescein diacetate succinimidyl ester ( CFSE; Invitrogen ) at a final concentration of 20 µM . Pellets were then centrifuged at 900 g for 10 min to remove the excessive stain and were diluted with pre-warmed ( 37°C ) serum-free culture medium . For each well of Caco-2 cells grown on coverslips in 24-well tissue culture plates , 1 . 25×107 parasites were added onto Caco-2 cell monolayer and incubated for 1 hour at 37°C . After incubation , cells were washed 5 times with sterile PBS and fixed with 2% formaldehyde for 30 min . The monolayers were then washed and stained with DAPI for 10 min and washed again . All monolayers were mounted onto glass slides using fluorescence mounting media ( VECTASHIELD ) prior to confocal microscopic examination ( Olympus Fluoview FV1000 , Olympus , Japan ) . Caco-2 monolayers were grown on Millipore transwell system until they reached confluency and tight junction maturation on day 21 . After confirmation of maturation by TER measurement , monolayers were co-incubated with parasite live cells for 24 h . Following co-incubation , epithelial and basolateral compartments were washed twice , followed by addition of 400 µl of warm ( 37°C ) HBSS at the basolateral compartments , and 200 µl of 100 mg/ml FITC-conjugated Dextran 4000 ( Sigma ) solution in HBSS to apical compartments . After 1 h at 37°C , 300 µl of HBSS was taken from basolateral compartments and was transferred to a black 96-well plate ( NUNC ) to estimate dextran–FITC flux across monolayers . Fluorescence was measured using an ELISA reader ( Tecan Infinite M200 ) at excitation and emission wavelengths of 492 nm and 518 nm respectively . Caco-2 cells were plated at a density of 2×104/cm2 on a 6-well culture plate and allowed to reach confluency . After treatments , cell monolayers were then rinsed three times with chilled sterile PBS ( pH 7·4 ) and lysed on ice for 40 min with 150 µl of ice-cold radioimmunoprecipitation assay ( RIPA ) buffer ( 150 mMNaCl , 1 . 0% Triton X-100 , 0 . 5% sodium deoxycholate , 0 . 1% SDS , 50 mMTris-HCl , pH 8 . 0 ) , including protease and phosphotase inhibitors for proteins extraction . After centrifugation at 16 , 000 g for 30 min at 4°C , the supernatant was collected for further analysis . Equal amounts of total protein were separated on 10% SDS-polyacrylamide gels and then transferred to a nitrocellulose membrane . After blocking for 2 h in PBS containing 0 . 1% Tween and 5% ( w/v ) skim milk , membranes were incubated overnight at 4°C , in primary antibody ( 1∶3000 , mouse anti-ZO-1 , anti-Occludin; Zymed ) . After five washes for 5 min each with PBS-T , the membranes were incubated for 1 h with horseradish peroxidase-conjugated secondary antibody ( 1∶2000 ) . After another five washes with PBS-T , the membranes were developed using enhanced chemiluminescence reagent ( Amersham , Princeton , NJ , USA ) to detect proteins . Nitric oxide cytotoxicity against Blastocystis was investigated by confocal microscopy ( Annexin-FITC/PI staining ) . Live Blastocystis were treated for 3 h with a 50 µg/ml concentration of GSNO ( S-Nitrosoglutathione ) , a nitric oxide donor . After drug exposure , the parasites were washed and re-suspended in Annexin V binding buffer ( BioVision ) . Fluorescein isothiocyanate ( FITC ) -labeled annexin V and propidium iodide ( PI ) ( BioVision ) were then added to the cell suspension at the ratio of 1∶100 . Imaging of cell suspensions was conducted using a confocal microscope ( Olympus Fluoview FV1000 , Olympus , Japan ) . Images were captured using Olympus Fluoview v . 1 . 6b . To study tight junction proteins in Caco-2 cells , Caco-2 cells were grown on Poly-L-lysine treated glass coverslips to 100% confluency . After treatment , monolayers of cells were rinsed three times with chilled sterile PBS ( pH 7·4 ) , fixed with 2% formaldehyde for 30 min , permeabilized with 0 . 1% Triton ×100 for 10 min , blocked with 10% FBS for 2 h at room temperature and then incubated with respective primary antibodies to ZO-1 and occludin overnight at 4°C . After five washes for 5 min each with PBS , the monolayer was then incubated with Cy3-conjugated secondary antibodies . Glass coverslips were then mounted onto clean glass slides using fluorescence mounting media ( VECTASHIELD ) and examined under confocal laser microscopy . To explore the spatial relationship between Blastocystis and Caco-2 monolayers after attachment , Caco-2 monolayers were grown to confluency on glass converslips and were then co-incubated for 1 h with live Blastocystis cells . After 1 h incubation , cells were washed 5 times with sterile PBS and fixed with 2% formaldehyde for 30 min , blocked with 10% FBS for 2 h at room temperature and then incubated with legumain antibody-mAb1D5 [35] , an in house murine IgM monoclonal antibody [61] , and secondary Alexa Fluor® 594 goat anti-mouse IgM ( red ) . Phalloidin-FITC ( green ) was used to label F-actin of Caco-2 cells and DAPI ( blue ) for nuclei . All monolayers were mounted on to glass slides using fluorescence mounting media ( VECTASHIELD ) before confocal microscopic examination . Mz , GSNO and NaNO2 were purchased from Sigma . Stock solutions of each compound to be tested were prepared fresh in DMSO . For drug sensitivity determination , stock solutions were diluted in prereduced Blastocystis medium and transferred to 96-well plates and pre-reduced again for 4 h . The final DMSO concentration was kept constant at 0 . 5% . The ANOVA test was used to confirm the statistical significance of the results . Correlation analyses between Mz resistance and fitness parameters ( i . e . attachment , permeability increase and nitric oxide tolerance ) were performed using the Pearson correlation method in GraphPad Prism 6 ( GraphPad Software Inc . , San Diego , CA , USA ) .
Given that Blastocystis induces barrier dysfunction in an in vitro model [36] and that clinical and animal studies suggest an induction of barrier defects by Blastocystis [2] , [8] , [33] , [62] , we investigated the variability in the ability of different parasite isolates to breach the epithelial barrier . Live parasites of all Blastocystis isolates used were applied apically on to Caco-2 cell monolayers differentiated on transwell inserts . Permeability changes caused by the parasites were analyzed by measuring the flux of a dextran–fluorescein isothiocyanate ( FITC ) probe across the intestinal epithelial barrier from the apical to the basolateral compartment . Exposure to ST-4 isolates WR-1 and S-1 did not change permeability significantly compared with that of the control ( Figure 1 ) . In contrast , all five ST-7 strains of Blastocystis could induce significant increase in epithelial permeability compared with the control and ST-4-treated monolayers ( p<0 . 01 ) . Within ST-7 , extensive variations in the ability to induce permeability increase were observed . Compared with the negative control monolayers , the fold increase in permeability to FITC-dextran in ST-7-infected monolayers were 16 , 11 . 8 , 14 . 5 , 5 . 5 , 4 . 2 , respectively , for isolates H , G , C , B and E ( Table 1 ) . Among them , C , G and H induced a significantly higher permeability increase compared with isolates B and E ( p<0 . 01 ) ( Figure 1 ) . The permeability change caused by isolate H was the most prominent and was more than three times higher than that by isolate E ( Table 1 ) . Altogether , it suggested intra- and inter-subtype variation in barrier disruptive activity of different Blastocystis isolates . Tight junction ( TJ ) complexes regulate epithelial barrier function [63] . Given that observed previously a ST-7 isolate of Blastocystis induced epithelial barrier defect by disruption of tight junction protein ZO-1 [36] , we proceeded to conduct an in-depth analysis of the changes in Caco-2 tight junction proteins occludin and ZO-1 after exposure to all of the seven strains of Blastocystis . Consistent with the permeability results , the profile of tight junction degradation shown by Western blot analysis differed from strain to strain and correlated with the degree of permeability increase caused by each strain ( Figure 2A&B ) . ST-4 treatments did not alter the two tight junction proteins significantly which might explain the insignificant permeability increase seen above ( Figure 1 ) . Expression of occludin , a transmembrane protein which plays a direct role in the paracellular permeability [64] , was significantly decreased compared with the control and ST-4 treatments by exposure to ST-7 isolates C , G , and H ( p<0 . 01 ) ( Figure 2A&B ) . But no changes in occludin were observed in ST-7 isolates B- and E-treated cells , which might contribute to the less pronounced permeability increase by B and E compared with C , G and H . The other tight junction protein ZO-1 , a multi-domain scaffold protein localized at the tight junction , however , showed a different pattern of disruption from that of occludin . All ST-7 strains induced a significant decrease in ZO-1 band intensity compared with the negative control , which might explain the prominent permeability increase in B- , E-treated cells compared with control monolayers; nevertheless , the degree of ZO-1 degradation by B and E was again significantly less than C , G and H ( p<0 . 01 ) ( Figure 2A&B ) , suggesting a more potent ability in disrupting ZO-1 by C , G and H among ST-7 isolates . In addition , immunofluorescence was performed to observe the distribution and integrity of the tight junction proteins . Indirect immunolabelling of occludin in control monolayers showed a bright and continuous band lining cell-to-cell contact regions ( Figure 2C ) . The staining pattern in cells treated by ST-4 ( WR-1 and S-1 ) and ST-7 ( B and E ) isolates did not differ from that of the controls . In contrast , infection of Caco-2 cell monolayers with ST-7 ( C and G ) led to focal disruptions and punctate concentration along pericellular junctions , while there was almost a complete loss of occludin in H-treated cells . As expected , ZO-1 proteins in cells after ST-4 infection appeared to be similar to those of control monolayers , where they showed a typical , continuous pericellular organization . Treatments with ST-7 ( B and E ) , however , resulted in both a reduction in ZO-1 intensity and reorganization of ZO-1 ( Figure 2D ) , as reported previously for ST-7 ( B ) -treated Caco-2 cells [36] , whereas exposure to ST-7 ( C , G and H ) isolates led to a complete degradation of ZO-1 at the apical junctions ( Figure 2D ) . In many parasites and other pathogens , adhesion is directly associated with the virulence properties of the strains [40] , [42] . Therefore , we tested whether there was a difference in the adhesion properties of different isolates and whether it correlated with the ability of the parasites to induce a permeability increase in intestinal monolayers . To detect and quantify adhesion of different Blastocystis isolates to host cells , an adhesion assay was developed using Caco-2 cell line . Both of the ST-4 strains , WR-1 and S-1 , showed negligible adhesion ( Figure 3A&B ) . Within ST-7 , isolates B and E adhered to host cells in a significantly lower number than C , G and H ( p<0 . 05 ) ( Figure 3B ) . From the merged image , it was clear that C , G and H parasites attached in large numbers ( Figure 3A ) . Isolate H , which induced highest epithelial permeability increase , was also the most adhesive to host cells ( Figure 3B ) . The results suggested an association between the level of attachment and the ability to induce permeability increase . An analysis of these two phenotypes showed a significant positive correlation , suggesting that adherence might play a role in virulence ( R2 = 0 . 8506 , p<0 . 01 ) ( Figure 3C ) . Even though isolates S-1 and E exhibited similar level of attachment ( Figure 3B ) , the former failed to induce a permeability increase in Caco-2 monolayers ( Figure 1 ) , suggesting that additional factors of isolate E might contribute to the permeability increase . Furthermore , we also provided confocal microscopic evidence of Blastocystis-induced pathological changes through parasite-enterocyte contact ( Figure 4 ) . Blastocystis ST-7 ( H ) preferentially adhered at the apical junction region ( Figure 4A ) . The percentage of parasites preferentially attached to the intercellular junctions was 80%±3 . 8% of all attached cells ( values are means ± SE ) . Parasites adhered intimately with host cells at the cell-cell junction site and induced an increase in actin polymerization ( Figure 4B&C ) . In addition , a loss of cellular symmetry in host cells was observed compared with the negative control ( which displayed an intact and normally organized epithelium ) ( Figure 4B ) , indicating adhesion-mediated intestinal epithelial injury . Moreover , we tested whether disruption of adhesion could restore the pathological alterations induced by Blastocystis . Results obtained with the most adhesive isolate H , showed that galactose effectively decreased H adhesion in a dose-dependent manner ( 67 . 7%±14 . 1% , 33%±0 . 3% of control at 50 and 100 mM , respectively ) ( Figure 5A ) ( p<0 . 01 for both values ) , while glucose , as a control sugar , did not have any significant effect on Blastocystis adhesion . Importantly , the ZO-1 tight junction protein was prevented from degradation by ST-7 ( H ) in the presence of galactose at 100 mM ( Figure 5B&C ) , reiterating the significant role of adhesion in mediating the pathogenesis of disease . As cysteine proteases ( CPs ) have been suspected to be virulence factors in Blastocystis and shown to play a role in inducing barrier compromise [36] , we compared the cysteine protease activity of all the seven Blastocystis isolates . CP activities in respective parasite lysates were determined using the azocasein assay . All tested strains tested showed significantly higher protease activity than the PBS control . Variation in protease activity was observed among different Blastocystis isolates ( Figure 6 ) . ST-4 isolates WR-1 and S-1 showed protease activity of 13 . 86±2 . 5 and 13 . 25±4 . 2 azocasein units . Except isolate G , most of the ST-7 strains showed significantly higher protease activity than ST-4 isolates ( p<0 . 01 ) , with 30 . 41±0 . 94 , 34 . 99±3 . 04 , 26 . 03±2 . 10 , 26 . 83±4 . 17 azocasein units for B , E , C , and H , respectively . Lysate of G isolate showed activity of 16 . 77±3 . 98 units , which was not significantly different from ST-4 strains . There was no correlation between the total CP activities and the virulence of each strain , suggesting that total cellular cysteine protease is not responsible for the variation in pathogenicity within ST-7 isolates in inducing intestinal barrier dysfunction . Previously , we reported that isolates B and E belonging to ST-7 were Mz resistant [49] . As both of them also exhibited impaired ability to adhere to host cells , we tested whether there was an association between attachment and Mz resistance in ST-7 , as seen in Giardia [54] . IC-50s of C , G and H isolates ( 2 . 98±0 . 97 , 3 . 63±0 . 96 , 1 . 05±0 . 32 µg/ml , respectively; Table 2 ) were significantly lower than the previously reported IC50s of isolates B and E ( p<0 . 01; Figure 7 ) , indicating an intra-subtype variation in Mz resistance in Blastocystis ST-7 parasites . Correlation analysis showed a significant negative correlation between the level of Mz resistance of an isolate and its ability to adhere to host cells ( R2 = 0 . 8908 , p = 0 . 0158 ) ( Figure 8A and Table 3 ) . Our results suggest that Mz resistance might entail an attachment defect in Blastocystis ST-7 strains . As we showed that the level of attachment was correlated with the ability to induce permeability increase ( Figure 3C ) , the impaired attachment in Mzr strains might indicate a less potent ability in inducing a permeability increase . Indeed , the correlation analysis between drug resistance and permeability increase showed a significant negative correlation ( R2 = 0 . 9644 , p<0 . 01 ) ( Figure 8B and Table 3 ) . Our results suggested that Mz resistance was negatively correlated with barrier disruptive ability of Blastocystis which in turn dampens its virulence . Most drug resistance mechanisms are associated with a fitness cost that is typically observed as a reduced growth rate [51] , [65] . To investigate the functional consequences of Mz resistance in Blastocystis , we examined the proliferative potential of Mzr and Mzs strains from growth curves assayed over 96 h . The growth rates of Mzr isolates were not always lower than Mzs isolates ( Figure S2 ) . Isolate E , being highly resistant to Mz , consistently grew more rapidly than Mzs isolates C and G , suggesting that drug resistance in Blastocystis ST-7 might not necessarily result in slower growth . Aside from adhesion , microorganisms also need to evolve means to counter the host immune system for effective colonization in the intestinal lumen [37] . Nitric oxide ( NO ) is a potent innate immune response against a range of pathogens and nitrosative stress prevents colonization of host tissues by parasites and bacteria susceptible to NO [66] , [67] . Previous studies [30] , [68] have already suggested NO induces cell death in Blastocystis . Thus , the ability of Blastocystis to tolerate nitrosative stress is likely to be important for its survival in the gut lumen . Previously , we observed that a Mzr strain of Blastocystis exhibited lower tolerance to nitric oxide toxicity compared with a Mzs strain [30] . Here , we tested the IC50s of NO donors against the Mzr and Mzs strains , to examine their ability to tolerate nitrosative stress . Interestingly , Mzs isolates C , G and H of Blastocystis ST-7 , all exhibited significantly higher IC-50s of GSNO ( 61 . 79±4 . 03 , 75 . 24±3 . 99 , 121 . 73±8 . 17 µg/ml , respectively; p<0 . 01; Figure S3A and Table 2 ) than Mzr isolates B and E ( 30 . 63±1 . 67 , 30 . 92±1 . 73 µg/ml , respectively ) . Therefore , Mzr strains were more susceptible to GSNO . An analysis again showed a significant negative correlation between the level of Mz resistance of an isolate and its ability to tolerate GSNO toxicity ( R2 = 0 . 7972 , p<0 . 05 ) ( Figure S3B and Table 3 ) . IC50s with another NO donor NaNO2 for all these strains exhibited similar trends to those with GSNO ( Table 2 ) . Taken together , the results suggested that , although Mz resistance helps Blastocystis survive stress from chemotherapy , it makes Mzr strains less capable of coping with nitrosative stress . Concomitantly , to determine the morphological changes resulting from nitrosative stress , isolates H and B were stained with propidium iodide and annexin V-FITC after exposure to NO [35] , [49] . Necrotic cells incorporated both PI and annexin V-FITC stain and cells undergoing programmed cell death bind annexin V-FITC alone . After treatment with 50 µg/ml concentration of GSNO , features of both necrosis and apoptosis were observed in the Mzr isolate B , whereas the Mzs isolate H were less affected ( Figure S3C ) . Compared with cultures of C , G and H , Mzr isolate B and E exhibited a significantly higher percentage of cells undergoing cell death ( 37 . 54% and 32 . 93% for B and E respectively; 12 . 38% , 10 . 2% , 7 . 21% for isolates C , G and H , respectively; p<0 . 01 ) ( Figure S3D ) , which also indicates that the parasite cell death induced by NO was predominantly by necrosis with a minority of cells dying by apoptosis .
Although it is one of the commonest eukaryotic organisms present in the alimentary tract of human and nonhuman hosts worldwide [69] , the debate about the pathogenicity of Blastocystis continues [3] , [21] , [70]–[72] , and arguments for and against its pathogenicity both abound [69] . The enigmatic role of Blastocystis has mainly been due to a lack of basic knowledge about its biology and convincing evidence of its pathogenicity [69] . A major obstacle or challenge is the potential for intra- and inter-subtype variation in Blastocystis pathogenicity [1] , [21] . Although previous research attempted to validate this argument by using animal models [32] , the number of strains used were limited and no consensus on the appropriate in vivo infection model was reached for Blastocystis; thus , the infection outcomes might be reflective of infectivity rather than pathogenicity . Our study , using a well-established in vitro system mimicking the host-pathogen interplay , provides , for the first time , a comprehensive analysis of different strains from two clinically relevant subtypes , and gives evidence that Blastocystis exhibited not only inter- but also intra-subtype variability in causing barrier dysfunction . Disturbances in epithelial barrier function are commonly associated with intestinal inflammatory disorders and have been reported in symptomatic cases of Blastocystis infection [2] , [4] , [62] , [73] . Thus , our findings might help explain the conflicting data concerning the inconsistency in reports of Blastocystis-induced intestinal inflammatory disorders . Besides , compared to other pathogens , Blastocystis requires a higher infectious dose to induce comparable damage ( Table 1 ) . Previous studies have also shown that greater Blastocystis parasite load shows higher pathogenicity [15] , [17] . Notwithstanding , some Blastocystis isolates ( e . g . , C and H ) could induce intestinal barrier compromise at a level comparable to accepted pathogens like some of the Giardia and Cryptosporidium strains ( Table 1 ) . Paracellular permeability was regulated by tight junctions , which seal the space between neighboring cells , generating an impermeable barrier between the epithelium and the extracellular environment , protecting deeper tissues from external aggressions including microbial infections [74] . However , this first line of defense against infection has become one of the most exploited gates for pathogens to access and colonize the host organism [74] . Enteric pathogens have developed a broad and complex range of mechanisms to subvert the host tight junction [75] . In general , key mechanisms identified to date include direct rearrangement or degradation of specific tight junction proteins , reorganization of the cell cytoskeleton , and activation of host cell signaling events [76]–[78] . In this study , we observed disruption of tight junctions ( TJs ) which mainly resulted from degradation in at least two TJ proteins: ZO-1 protein , which is linked to the cytoskeleton and plays a pivotal role in the TJ architecture [79] , and occludin , which is important in maintaining the integrity and barrier function [64] , [80] . Previously , we reported that Blastocystis phophorylates MLC and leads to cytoskeletal and ZO-1 rearrangement [36] . Indeed , we also observed myosin light chain phosphorylation and TJ rearrangement in B- and E-treated Caco-2 cells besides degradation of ZO-1 protein ( Figure S4 ) . However , for Blastocystis isolates C , G , and H , although MLC was phosphorylated , degradation of TJs appeared to be the main mechanism , indicating the strategies utilized by Blastocystis to induce permeability increase are likely to be diverse and different isolates may have developed different major mechanism to compromise barrier function . In our previous study , the observation that ROCK inhibition did not completely rescue the increase in the epithelial permeability induced by the parasite [36] also suggested that Blastocystis utilizes more than one mechanism to breach the epithelial barrier . Epithelial attachment is an important factor determining the persistence and virulence of luminal pathogens [38] , [39] . Indeed , in our study , pathogenicity of Blastocystis was also found to be correlated with their ability to attach to epithelial cells . Importantly , inhibition of adhesion ameliorates the parasite's pathogenic effects on degradation of ZO-1 tight junction protein . How adhesion leads to intestinal barrier dysfunction needs to be investigated further . Giardia intestinalis was shown to produce harmful substances as a result of the host-pathogen contact [81] . Several enzymes from the secreted products have been identified in G . intestinalis , which are suggested to facilitate effective parasite adhesion and colonization of the human small intestine [82] , [83] . Whether the adhesion of Blastocystis with host cells also enables particular enzymes to be released and to participate in pathogenesis would be interesting to investigate . The Blastocystis genome has been available and in silico analysis of the ST-7 secretome predicted 75 putative secreted proteins , some of which may have a direct connection with pathogenicity [34] . Recently , two cysteine proteases ( legumain and cathepsin B ) have been characterized in the ST-7 culture supernatant; these enzymes showed proteolytic activities by gelatin zymograms [84] . However , whether these secreted proteins act on intestinal cells and disturb gut function has not been studied . It would be interesting to investigate whether adhesion of Blastocystis to the epithelium might trigger the parasite to actively produce virulence factors that possibly activate signaling cascade leading to tight junction disruption . The observation of preferential adherence of cytopathic Blastocystis strains to intercellular junctions also raised the interesting question as to whether the preferred attachment at junctional area facilitates tight junction degradation by Blastocystis . Parasites such as Giardia intestinalis and Entamoeba histolytica have distinct virulent and non-virulent strains that may be attributable to qualitative and quantitative variation in their cysteine proteases activity [60] , [85] , [86] . Blastocystis cysteine proteases are also implicated in the activation of NF-κB in colonic epithelium , leading to an upregulation of the pro-inflammatory cytokine IL-8 [87] . Our most recent study [36] has also shown that cysteine proteases could cause human epithelial barrier compromise . Indeed , isolate E , even though being the least adhesive in ST-7 , has yet the highest cysteine protease activity , which might explain the permeability increase observed in E-infected epithelium; while isolate S-1 , with equivalent level of attachment to E , and at the same time the lowest CP activity , failed to induce any increase in intestinal permeability . Interestingly , the highly adhesive G isolate , though with similar CP activity to ST-4 isolates , could induce prominent epithelial barrier disruption , highlighting the important role of adhesion in mediating intestinal pathology . Taken together , both adhesion and CP activity might contribute to Blastocystis–induced barrier dysfunction . ST-4 isolates , low in both adhesion and CP activity , appeared to be avirulent for human intestinal monolayer , whereas all ST-7 isolates studied are capable of infection because of being either adhesive or exhibiting high CP activity , or both . Moreover , adhesion is a major determinant in the pathogenicity of a strain . The results of our study provide evidence that the pathogenesis of blastocystosis is complex and parasite adhesion and cysteine proteinases might be important virulence factors . Specific virulence factors and their mechanisms remain to be explored in depth . Besides the observation that Blastocystis ST-7 exhibited intra-subtype variation in epithelial attachment , we also observed an association between epithelial attachment and Mz susceptibility . Mzr isolates in ST-7 exhibited a defect in epithelial attachment . How Mz resistance would affect Blastocystis adhesion to host cells remains to be addressed . Drug resistance has been shown to modulate adhesion in some bacterial and parasitic agents , such as Escherichia coli , Giardia , and Stenotrophomonas [54] , [88]–[94] . Tejman-Yarden et al . [54] showed that in Giardia lamblia , epithelial attachment was linked to glucose metabolism . Glucose promoted parasite attachment and Mzr lines were shown to consume less glucose than their parental Mzs lines , providing a possible explanation for the attachment defect observed in Mzr lines . In Blastocystis , we also measured the glucose consumption of all seven strains used in the study . However , their respective glucose consumption rate did not correlate with the level of Mz resistance ( Figure S5 ) , suggesting that glucose metabolism may play a different role in Blastocystis from that in Giardia . In addition , it was suggested that specific variable surface proteins ( VSPs ) might play a role in Giardia adherence and that to analyze the impact of Mz resistance on the VSP pattern may reveal new insights into the functions of specific VSPs in parasite attachment [54] . In Blastocystis , whether the decreased adhesion observed in Mzr strains indicated that Mz selection negatively affected the synthesis or expression of surface adhesion molecules might be worth exploring . On the other hand , little is known about the nature of adhesion of Blastocystis to host cells and many basic questions ( such as what molecules mediate parasite-host adhesion and what factors influence the process ) need to be addressed . Further dissection of the mechanisms of Blastocystis attachment to host cells will shed new light on the association between Mz resistance and attachment . Mzr strains not only exhibited a defect in attachment , they also showed impaired NO tolerance in Blastocystis ST-7 . NO is an important aspect in human innate immunity and limits microbial persistence in the gut by inducing cell death and preventing their growth and encystations [66] . Therefore , the ability to tolerate nitrosative stress would be an important aspect for parasite survival . Our study suggests that Mz resistance might limit its survivability to host innate immune response , such as coping with nitrosative stress . How Mz resistance affects the susceptibility to NO needs to be investigated further . Development of Mz resistance in microorganisms is a multifactorial process and the mechanisms are diverse [95] . In Giardia and Trichomonas , Mz resistance usually develops due to impairment of Mz activation enzymes in the parasite . For example , resistance to Mz has been associated with down-regulation or decreased activity of pyruvate: ferredoxinoxidoreductase ( PFOR ) , which activates the 5-nitroimidazole ( 5-NI ) pro-drugs to the toxic radical states inside the parasite in conjunction with the electron acceptor ferredoxin [54] , [96] , [97] . These Mz activation enzymes are oxygen-sensitive and are also important components of parasite anaerobic respiratory system . Anaerobic respiratory enzymes also play an essential role in free radical ( e . g . , H2O2 or NO ) scavenging and , if these enzymes are impaired , as expected in Mz resistant parasite strains , it would lead to decreased free radical tolerance in these isolates [98] , [99] . However , the mechanism of Mz activation has not been studied in Blastocystis , although PFOR and other oxidoreductase enzymes are present in the organism [34] , [100] . Further investigation of Mz activation and/or resistance mechanisms as well as NO-scavenging mechanisms in Blastocystis would be necessary to elucidate the association between NO susceptibility and Mz resistance . Recent studies of Giardia and Trichomonas suggested that Mz was activated by the nitroreductase activity of flavin-dependent thioredoxinreductase and Mz resistance was associated with decreased NADPH oxidase activity [95] , [101] . A flavohemoglobin ( FlavoHb ) , which plays a pivotal role in NO detoxification , has been characterized recently in Giardia intestinalis [102] . Interestingly , other studies have suggested flavoHb also demonstrates NADPH oxidase activity [103] . Furthermore , flavoHb seems to be important in the pathogenesis of disease [104] , [105] . It would be intriguing to explore whether flavoHb or its homolog exists in Blastocystis and whether it plays any function in Mz resistance , NO-scavenging or pathogenesis . In conclusion , our study , for the first time , showed extensive intra- and inter- subtype variability in pathogenicity in Blastocystis ST-4 and ST-7 in vitro . The existence of both pathogenic and apathogenic isolates might help explain the disparity in symptoms of among patients with Blastocystis infection . The modes of pathogenesis of blastocystosis are diverse , and the range of clinical symptoms observed is unlikely to be attributed to a single virulence factor or pathogenic mechanism . Moreover , our study indicates that adhesion is an important biomarker for the pathogenicity of the parasite and subsequent outcome of infection/disease , and that adhesion can be impaired by Mz resistance . Interestingly , the five ST-7 strains tested here were morphologically indistinguishable from each other , but varied significantly in many aspects of their pathobiology , including attachment and drug resistance . A deeper analysis and systematic comparison of the genome and protein expression profiles among these Blastocystis strains could aid in the identification and analysis of the key factors associated with the pathobiology of blastocystosis .
|
Since it was first described more than a century ago , the question as to whether the protistan parasite Blastocystis causes disease or is a commensal of the human gut still remains unresolved . Strain- or subtype-dependent variability in Blastocystis pathogenicity has been proposed to contribute to this controversy . Currently , the factors determining this variation are unknown . For seven strains from two clinically relevant Blastocystis subtypes , we evaluated their ability to adhere to human intestinal epithelium and to disturb barrier function . We showed that the more adhesive strains exhibited greater pathogenicity . We also observed an inverse correlation between adhesiveness and drug resistance , suggesting that drug resistance might compromise the fitness of the parasite . This is the first study highlighting the important role of adhesion in Blastocystis pathogenesis . We conclude that the extensive variation in Blastocystis pathogenicity is a plausible factor contributing to the disparate outcomes of Blastocystis infections .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"blastocystis",
"hominis",
"infection",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"emerging",
"infectious",
"diseases",
"pathology",
"and",
"laboratory",
"medicine",
"host-pathogen",
"interactions",
"medical",
"microbiology",
"gastroenterology",
"and",
"hepatology",
"microbial",
"pathogens",
"biology",
"and",
"life",
"sciences",
"microbiology",
"gastrointestinal",
"infections",
"parasitic",
"diseases",
"parasitology",
"parasite",
"physiology",
"pathogenesis"
] |
2014
|
Intra-Subtype Variation in Enteroadhesion Accounts for Differences in Epithelial Barrier Disruption and Is Associated with Metronidazole Resistance in Blastocystis Subtype-7
|
Mounting evidence shows mammalian brains are probabilistic computers , but the specific cells involved remain elusive . Parallel research suggests that grid cells of the mammalian hippocampal formation are fundamental to spatial cognition but their diverse response properties still defy explanation . No plausible model exists which explains stable grids in darkness for twenty minutes or longer , despite being one of the first results ever published on grid cells . Similarly , no current explanation can tie together grid fragmentation and grid rescaling , which show very different forms of flexibility in grid responses when the environment is varied . Other properties such as attractor dynamics and grid anisotropy seem to be at odds with one another unless additional properties are assumed such as a varying velocity gain . Modelling efforts have largely ignored the breadth of response patterns , while also failing to account for the disastrous effects of sensory noise during spatial learning and recall , especially in darkness . Here , published electrophysiological evidence from a range of experiments are reinterpreted using a novel probabilistic learning model , which shows that grid cell responses are accurately predicted by a probabilistic learning process . Diverse response properties of probabilistic grid cells are statistically indistinguishable from rat grid cells across key manipulations . A simple coherent set of probabilistic computations explains stable grid fields in darkness , partial grid rescaling in resized arenas , low-dimensional attractor grid cell dynamics , and grid fragmentation in hairpin mazes . The same computations also reconcile oscillatory dynamics at the single cell level with attractor dynamics at the cell ensemble level . Additionally , a clear functional role for boundary cells is proposed for spatial learning . These findings provide a parsimonious and unified explanation of grid cell function , and implicate grid cells as an accessible neuronal population readout of a set of probabilistic spatial computations .
Mammals use probabilistic computations to perceive noisy and ambiguous sensory inputs [1–5] . It seems likely that learning an internal model of a noisy sensory environment should follow similar statistical inference principles [4] . While solid behavioural evidence [1–5] and mounting in vivo evidence [3 , 4] support probabilistic sensory perception , in vivo evidence is lacking for probabilistic learning [4 , 5] . It is virtually unknown how any probabilistically learned neural model of the world may look through neurophysiological recordings . The mammalian hippocampal formation is heavily implicated in spatial learning [6–9] . Grid cells within the hippocampal formation tile Euclidean space in a repeating firing pattern , thought to provide a spatial metric [7–11] . Both theoretical and experimental evidence suggest that grid cells may be used for path integration ( PI ) via integration of self-motion estimates [8 , 10 , 12–14] . However , all PI systems suffer from cumulative error [15 , 16] necessitating frequent corrections [17–22] . In darkness [10 , 13 , 23 , 24] , fusion of sensory and learned information is necessary to maintain spatially-stable grid cell responses [17 , 18 , 25] . Theoretically , learned boundary information is sufficient to correct cumulative PI errors in darkness [17 , 18] . Consistent with theory , boundary cells have been found to fire along arena boundaries [26–29] , coexist with grid cells in the hippocampal formation , and provide a plausible neuronal substrate to encode boundary information [17–20] . However , it is unclear how grid and boundary information contribute to spatial learning , or how their responses may be altered by learning . Currently , no realistic learning model can unify grid and boundary cell activity for learning or localization in light and dark conditions . Darkness poses a formidable challenge by limiting inputs to noisy self-motion and intermittent boundary contacts , neither being location-specific . The approach of approximating spatial learning by assuming error-free PI by grid cells [20 , 22] bypasses the fundamental problem of SLAM ( simultaneous localization and mapping ) [30–32] , overlooking how cumulative errors [15 , 16] impair spatial learning and undoubtedly shaped the evolution of spatial cognition . Spatial learning models which rely on vision [21 , 33] do not generalize to explain stable grid fields in darkness [10 , 13] . Additionally , the nature of learned spatial information must affect the diverse grid cell responses caused by arena manipulations , including grid rescaling [34 , 35] and grid fragmentation [36 , 37] , but whose relationship to learned spatial representations remain unclear . Here , a new learning model adapted from probabilistic SLAM [30–32] is proposed which explains how information encoded by grid and boundary cell populations may be simultaneously learned , recalled and stabilized . This model takes realistic noisy inputs , represented by plausible neuronal codes using rate-coded ( non-spiking ) grid and boundary cell models , and carries out probabilistic information fusion algorithmically . Grid and boundary cell responses are modulated recursively through information fusion , during both learning and recall . Expected neuronal responses are characterized and compared to rodent electrophysiological recordings across diverse conditions .
A probabilistic grid and boundary cell spatial information fusion model ( SIFM ) was developed ( Figs 1A and S1 , Methods , S1 . 1 Text ) based on a Rao-Blackwellized particle-filter [30–32] . This algorithmic implementation of information fusion continually updates a sample of possible positions , orientations , and environmental layouts ( maps ) , primarily using self-motion and boundary information . During a learning session , boundary information is associated with position information incrementally to build a distribution of possible maps corresponding to the possible paths defined by the dynamic distribution of positions and orientations . At each step boundary prediction error feeds back to modulate the distribution of positions , orientations and maps . SIFM is the first demonstration that information contained in grid and boundary codes of the rodent hippocampal formation may be sufficient to implement recursive probabilistic information fusion . In SIFM , noisy self-motion cues update grid cell responses ( Fig 1B and 1C , S1 . 1 . 2 Text ) , encoding a distribution of position and orientation ( pose ) , spanning eight modular grid scales estimated from rodent data [35] . Noisy boundary cues are encoded by boundary cells covering twelve tuning directions and nineteen tuning distances ( Fig 1D and 1E ) . Eleven long-range boundary cell tuning distances were inferred previously from hippocampal place cell data [38] , while eight additional short-range tuning distances enable boundary detection in darkness , consistent with medial entorhinal boundary cell properties [26] . From self-motion and boundary cues experienced along random trajectories , each multi-modular grid code learns an association map ( weights to boundary cells , Figs 1A and S1 , S1 . 1 . 6 Text ) , which together determines predictive boundary cell responses . Each multi-modular grid code is assumed to maintain the same relative phase across grid scales ( modules ) despite continual changes in position . Note that although a self-corrective mechanism could potentially maintain relative grid phase across modules [39] , partial decoupling and hence independent operation of some grid modules [35] does not prohibit SIFM function ( described later under ‘Three novel tests of probabilistic learning’ ) . During learning and recall , boundary prediction error recursively modulates the grid code distribution ( S1 Fig , S1 . 1 . 7 Text ) . Sensory noise was based on previous analyses of published rat experiments [17 , 18 , 40 , 41] , while trajectory characteristics and sensory information were constrained by rat physiology and published experimental designs ( see S1 . 1–S1 . 3 Text for details ) . In arenas similar to published grid cell experiments [10 , 11 , 13 , 26 , 34 , 35] , stable hexagonal grid patterns were seen ( gridness index mean ± SD = 1 . 23 ± 0 . 07 in 1 m square arena , n = 4 , 000; 1 . 17 ± 0 . 11 in 1 m circular arena , n = 4 , 000 ) consistent with rodent grid cells [29 , 42 , 43] ( Figs 2 , S2 and S3 , S1 Table ) . Like pure rodent grid cells [42] , probabilistic grid cells were directionally-insensitive ( directional information content < 0 . 1 bits/spike ) . Two distinct types of grid phase noise exist in SIFM , one arising from self-motion cues which corrupts the estimate of pose and is correlated between all grid cells , the second arising intrinsically in grid codes and is independent between grid codes but correlated across modules within each grid code . This second type of phase noise is used opportunistically to compensate for the information loss due to self-motion errors [17 , 18 , 30–32] ( Figs 1A and S1 ) . By virtue of its independence between grid codes , compensatory phase noise serves to sample the phase space to keep track of pose uncertainty , which is critical for successful information fusion . Without compensatory phase noise , grid cell responses were perfectly coupled , carrying redundant information , resulting in pure PI ( Fig 2 rows 2 and 4 ) and unstable grid fields ( gridness index mean ± SD = -0 . 05 ± 0 . 27 and -0 . 10 ± 0 . 29 for the square and circular arena , respectively ) . Notably , SIFM can also learn using spatially irregular grid codes ( S4 Fig ) , in keeping with recent reports of anisotropic but stable grids [44] . A unique feature of SIFM are predictive boundary cells downstream of grid cells ( Figs 1A and S1 , S1 . 1 Text ) , which operate together with previously described sensory boundary cells [27 , 38 , 45] . Successful learning resulted in predictive boundary cell responses similar to rodent boundary cells [26 , 27 , 29] ( Figs 2C , S2 and S3 ) . However , absence of compensatory grid phase noise led to spatially-dispersed associations learned between grid and boundary cells ( Fig 2A rows 2 and 4 ) , because grid responses themselves were spatially-dispersed ( Fig 2B ) . Hence stability of predictive boundary fields ( Fig 2C ) depends on grid stability ( Fig 2B ) , so that inhibiting probabilistic learning by removing compensatory grid cell phase noise also prevents stable predictive boundary fields being established . Surprisingly , approximately one third of predictive boundary cells were unclassified using the border score [26] ( b<0 . 5 , S5 Fig ) raising the possibility that many predictive boundary cells do not encode boundary information . Since some rodent boundary cells are active parallel to , but disjoint from , a boundary [26 , 27] , consistent with model predictions [27 , 38 , 45] , a new hypothesis-based classification procedure was developed ( S1 . 3 . 6 Text ) which correctly classified predictive boundary cells , showing that boundary information is encoded ( S5 Fig ) . Finally , some predictive boundary cells showed fractional activity along a wall ( S2C and S3C Figs ) , similar to some rodent boundary cells [26] . Fractional activity arose from a wall being oblique to the tuning direction ( bottom row of S2 and S3 Figs ) . Rodent grid cell rate maps are immediately stable in darkness in a novel room ( Fig S7b of [10] ) , showing that grid stability and alignment do not require visual cues or familiar room landmarks . Potentially , PI is used to track location , but cumulative sensory errors cause failure within 1 to 2 minutes [17 , 18] , an order of magnitude less than rodent grid cell results [10] . Similarly , learning the new environment using PI would also fail ( Fig 2 ) because PI errors accumulate if left uncorrected leading to increasing uncertainty and unstable grids [12 , 46] . However , prior exposure to an identical arena in a different room [10] could plausibly have established an equivalent map in memory , which is additional information required for information fusion and grid stabilization . With prior exposure to a geometrically-equivalent arena , probabilistic grid fields are also immediately stable in darkness , despite initial disorientation ( Fig 3A Novel / total darkness ) . Also consistent with rodent results , probabilistic grid fields remained stable in light ( Fig 3A Light ) , and in a second darkness session ( Fig 3A Dark ) . Re-orientation was possible due to the arena’s local rotational asymmetry , allowing the fusion of self-motion and boundary information to converge on one of four possible solutions [18] . Local arena asymmetry was shown previously to improve localization due to the limited number of rotationally equivalent poses which are consistent with sensory information [17 , 18] . Hence the coupling between probabilistic grid cells and predictive boundary cells accounts for the persistence of rodent boundary fields in darkness ( Fig 3A rows 4 and 5 ) [27] . Since circular arenas possess no local or global rotational asymmetry [18] , perhaps stable grid fields in darkness ( Fig S7a of [10] ) cannot be explained by probabilistic models . However , SIFM slowed spatial destabilization sufficiently to allow grid fields to persist ( Fig 3B ) . Bin-wise correlations between probabilistic grid maps in light and darkness were almost identical to rats ( mean ± SEM , n = 33: for Light-to-Dark , rSIFM = 0 . 50 ± 0 . 03 , rrat = 0 . 50 ± 0 . 03 , P = 0 . 97 , t32 . 3 = -0 . 04; for Dark-to-Light , rSIFM = 0 . 53 ± 0 . 03 , rrat = 0 . 53 ± 0 . 03 , P = 0 . 87 , t32 . 3 = -0 . 17; Welch’s t-test ) , showing that persistence of grids in darkness may be largely explained by the fusion of self-motion and learned boundary information . Hence both the emergence and persistence of stable grids in darkness is explained by the same probabilistic model . Grid patterns compress or stretch along the resized dimension of a test arena [34 , 35] , but the mechanism is poorly understood . One model assumed ideal associations between visual features and grid cells to reset grid activity [22] , deforming grid patterns in resized arenas . However , grids rescaled heterogeneously over the arena ( Figs 7A , B of [22] ) in contrast to rodent grids [34 , 35] , because dominant visual features locally anchored the model grid cell’s activity . A more recent model assumed place cells , driven by boundary vector cells , reset grid cell activity [19] . Both models were heavily biased towards boundary information , thus preserving the same grid peaks despite arena resizing , differing from rodent grid cell rate maps showing partial loss or gain of grid peaks [34] . Conversely , a particle filter model underestimated grid rescaling when boundary information was used infrequently [17 , 18] , due to the dominance of PI . In contrast , probabilistic grids rescaled partially , nearly identical to rodent data ( Fig 4 ) . Like rodent grid cells , probabilistic grids rescaled isotropically , leading to partial loss or gain of grid peaks at boundary edges ( e . g . , circled grid peaks in Fig 4 ) . Across twelve rescaling dimensions , rescaling magnitude was almost identical between probabilistic and rodent grid cells ( Fig 4E ) . Notably , probabilistic grid rescaling in resized arenas was not based on a resizing-dependent change in the velocity signal [12] , or altered oscillatory properties of grid cells . Instead , altered oscillation frequency caused omnidirectional probabilistic grid rescaling ( S6 Fig ) in an unchanged arena , and compatible with rodent grid expansion in novel environments [47] . In the latter , omnidirectional rescaling of probabilistic grids was caused by a change in the overall speed signal gain in the novel but geometrically equivalent arena . In contrast , the speed signal in rescaled arenas was unchanged , and grid rescaling was SIFM’s probabilistic solution to the conflict between learned and current environmental boundary information . Although probabilistic grid patterns varied substantially with either arena geometry or oscillatory dynamics , grid stabilization depended critically on boundary information , contrary to previous argument [43] . Instead , probabilistic grid rescaling resulted from the competition between boundary and self-motion information . In resized arenas , some grid codes closely matched PI position but not boundary code , and some vice versa . High boundary prediction error suppressed the PI-matching grid codes , biasing towards grid codes which better matched the boundary input ( S1 Fig , S1 . 1 Text ) , thereby gradually pulling the grid code distribution away from the PI estimate . Simultaneously , coupling of predictive boundary cells to grid cells caused predictive boundary fields to shift along the rescaling dimension ( S7 Fig ) , distinct from sensory boundary cells whose response distance to the boundary is fixed [38 , 45] . The former is a novel prediction of predictive boundary cells . The spatial regularity and invariance of rodent grid cell responses are consistent with low-dimensional attractor properties [8 , 12 , 43 , 48] . Local grid cell pairs showed highly correlated activity , adjusting for 2D translation , even when grid parameters were experimentally altered , suggesting strong intrinsic coupling of grid cell activity [11 , 43] . For example , observed high inter-grid correlation cannot be explained simply by the stability of individual grid properties , since those properties change significantly with experimental manipulations such as arena resizing . Since information fusion in SIFM specifically requires partially uncoupled grid cells to track uncertainty in grid phase ( e . g . , Fig 2 ) , perhaps probabilistic learning and recall ( S1 . 1 . 8 Text ) cannot explain attractor-like properties of rodent grid cells . To clearly test the strength of functional coupling between probabilistic grid cells , two recall tests were used with identical initial conditions ( Fig 5A ) . Learning was ceased during recall tests to prevent the possibility that a dynamically changing spatial memory may contribute to session-specific correlations in grid response . Hence , if there is stronger correlation within a grid cell between trials than between grid cells in the same trial , then correlation of between-cell activity may simply reflect between-trial stability [43] . Like rodent grid cells , the opposite is true for probabilistic grid cells whose properties were more stable between cells within a trial , than in the same cell between trials ( Fig 5B ) . The same analysis was repeated using an arena resizing series to determine if gross changes in grid parameters may weaken the functional correlation between grid cells ( S8 Fig ) . Again , grid parameters co-varied between cells across conditions , showing a nearly identical pattern to rodent grid cells up to a scale factor . Like rodent grid cells [43] , probabilistic grid cell parameter ratios also varied with omnidirectional rescaling in novel environments , with the changes virtually identical between all cell pairs within each session ( S9A Fig ) . Additionally , rodent grid cells show small phase changes across different test sessions in the same arena , which exceed the variability of between-cell phase offsets across the same sessions ( e . g . , Fig 2 of [43] ) . Since probabilistic grid peak locations depend on learned information , perhaps grid drift is negligible between recall sessions since learned information is unchanged . Instead , the magnitude of within-cell phase drift is significantly larger than the magnitude of change in between-cell phase offsets across two independent recall trials ( S9B Fig , P = 3 . 3 × 10−267 , z = 34 . 9 , Wilcoxon rank sum test between the Between-cell and Within-cell magnitude distributions , n = 2 , 000 ) . Similarly , the magnitude of within-cell phase drift is also larger than the phase offset of phase-matched grid cell pairs within each recall trial ( P < 10−300 , z = 42 . 5 , Wilcoxon rank sum test between the Within-cell magnitude distribution and phase offset distribution from Recall trial 1 , n = 2 , 000; P < 10−300 , z = 44 . 6 , Wilcoxon rank sum test between the Within-cell magnitude distribution and phase offset distribution from Recall trial 2 , n = 2 , 000 ) . These results demonstrate that probabilistic grid cells exhibit correlated phase changes across independent test sessions , similar to rodents . Notably , learned information was constant and equivalent between probabilistic recall sessions , suggesting that phase drift may primarily be a function of stochastic processes such as random movement and noise . To determine if momentary system dynamics also display attractor character analogous to temporally-averaged grid patterns , the response of 200 phase-matched probabilistic grid cells were examined . As expected from momentary functional coupling , response time series between all grid cells within a scale module were correlated , orbiting an attractor in activity space ( unit diagonal in Fig 5C ) . Recall localization tests were performed in darkness to exclude compass stability as a possible explanation for the high correlation between cells ( using arena 1-fold rotational symmetry for global orientation [18] ) . In contrast , there was a clear reduction in the activity correlation between grid cells during non-probabilistic learning or recall ( Fig 5D ) . Here , non-probabilistic learning was identical to probabilistic learning in all respects except for the loss of useful feedback from boundary prediction error thereby specifically preventing the probabilistic fusion of self-motion and boundary information while preserving the distributed estimate of pose and associative learning ( S10 Fig ) . Initially disoriented , the perpendicular distance between the activity state and the attractor reduced and then stabilized during probabilistic recall , but remained high during non-probabilistic recall ( Fig 5E ) . Once localized , correlation between intra-modular grid cells remained high ( Fig 5F , mean r > 0 . 8 after the first minute ) for probabilistic grid cells , but not for non-probabilistic grid cells ( Fig 5G , mean r < 0 . 02 ) , showing that successful localization was marked by strongly correlated grid cell activity . Taken together , these results demonstrate that probabilistic grid cells exhibit attractor properties similar to rodent grid cells , and that these properties specifically require probabilistic information fusion . Furthermore , attractor properties should be evident in darkness , initially disoriented , and on a momentary basis . SIFM also provides support that oscillatory and attractor dynamics may be complementary [19 , 49] . Notably , attractor-like SIFM grid cell correlations arise from shared sensory inputs and convergence in learned information rather than direct coupling between grid cells , consistent with recent noise correlation analysis of rodent grid cells suggesting that relatively little spike train correlation is attributable to direct synaptic connections between cell pairs [50] . Hexagonal grid patterns are lost in hairpin mazes [36 , 37] . Resulting grid cell rate maps fragment , repeating across alternating maze arms , acquiring a dependence on global running direction [36] . It remains unclear why inserted barriers alter grid cell activity in this particular way . It was hypothesized that grid maps fragment into multiple submaps , interconnected across hairpin turns [36 , 37] . If so , probabilistic grids should not fragment since only a single map is learned ( e . g . , Fig 2 ) , and no specific mechanism exists to connect or switch between maze arms . Surprisingly , probabilistic grid cell rate maps do fragment , alternate across arms in a hairpin maze ( Fig 6A ) , and depend on global running direction ( Fig 6C ) , similar to rats ( Fig 6B and 6D ) . Arm-arm correlation matrices showed similar checkerboard patterns for individual grid cells ( Fig 6C and 6D ) and for a population ( Fig 6E and 6F ) . Correlation between matrices from probabilistic and rat grid cells were high ( easterly , r = 0 . 92 , P = 9 . 9×10−43; westerly , r = 0 . 95 , P = 1 . 8×10−50; element-wise correlation of population correlations of all unique arm-arm pairs ) , but dropped if the global running direction was mismatched ( probabilistic easterly vs rat westerly , r = 0 . 82 , P = 2 . 0×10−25; probabilistic westerly vs rat easterly , r = 0 . 86 , P = 4 . 2×10−31 ) , suggesting rate maps are more complex than repeating submaps across alleys . Like rat grid maps ( Fig 7A ) , correlations between arms with the same local running direction were higher than for arms with: opposite local running direction ( P = 3 . 5x10-91 , z = 20 . 3 , Wilcoxon rank sum test ) ; the same local running direction but randomly shuffled bins ( P = 4 . 8x10-132 , z = 24 . 5 ) ; opposite running direction reflected along the north-south axis ( P = 9 . 0x10-112 , z = 22 . 5 ) ; opposite global running direction ( east vs west , P = 7 . 5x10-125 , z = 23 . 8 ) . The median arm-arm correlations of directionally-insensitive rat grid cells ( n = 73 ) were within the 95% resampling C . I . of probabilistic grid cells ( Fig 7B; 104 resamples of n = 73; different local direction , rat median r = 0 . 09 , 95% C . I . = 0 . 09 to 0 . 22; same local direction , rat median r = 0 . 67 , 95% C . I . = 0 . 61 to 0 . 74 ) . Perhaps grid fragmentation arose from the complex interplay between cumulative error , altered path structure in hairpin mazes , and arena geometry [15 , 17 , 18] , not the inserted maze walls per se[36] . If so , grid maps should still fragment in an open arena if hairpin-like paths are followed . Instead , both probabilistic and rat grid cells ( Fig 7C and 7D ) preserved their hexagonal grid patterns along virtual hairpin ( VH ) paths [36] . It should be noted that rats were trained progressively to run to successive turn locations along an approximate virtual hairpin path in an open arena ( Fig 7D ) , whereas SIFM grid cells were tested along an ideal hairpin path without further training ( Fig 7C ) . Bin-wise correlations were indistinguishable between rat and probabilistic grid maps for open arena and: hairpin maze , virtual hairpin maze , and a second open arena ( Fig 7E , P > 0 . 4 for all comparisons ) . Contrary to the multiple submap hypothesis [36 , 37] , probabilistic learning resulted in a single map ( Fig 7F ) . Multiple geometrically-similar alleys were partially compressed , reducing the east-west spatial extent while preserving the north-south extent . Incomplete compression caused easterly and westerly global runs to use non-identical map information , causing directional-dependence ( Fig 6C ) . Compression was reduced using semi-transparent maze barriers ( Fig 7F , ST ) , resulting in a hybrid between arm-arm repetition and global tessellation patterns . The semi-transparent maze was modelled by including perimeter boundary inputs to boundary cells at all times to provide global arena geometry information , in addition to immediately adjacent walls . Semi-transparent walls were assumed to make perimeter walls visible to the rat . Consequently , arms with similar running directions remained correlated ( P = 1 . 4x10-34 , z = 12 . 3 , Wilcoxon signed rank test ) but reduced compared to opaque walls ( P = 8 . 4x10-65 , z = -17 . 0 , Wilcoxon rank sum test ) , similar to rats [36] . This shows that global boundary cues can partially disambiguate repetitive local arena structure . To further investigate the underlying cause of map compression during probabilistic learning , two hypotheses were tested . First , perhaps the resolution of the association map was too low to reliably learn the hairpin corridor structure , leading to map compression . Probabilistic learning trials were repeated using 4-fold and 0 . 25-fold association map resolution , but both map compression and grid fragmentation persisted ( S11A and S11B Fig , S2 Table ) . Second , perhaps high positional uncertainty caused boundary cues to be more heavily weighted than self-motion cues during probabilistic information fusion , leading to ambiguity between adjacent corridors . If so , a sufficient reduction in self-motion noise should recover the hairpin structure during learning , and rescue the hexagonal tessellating grid pattern , which was indeed the case ( S11C and S11D Fig , S2 Table ) . Taken together , these results suggest that grid fragmentation and map compression are due to the interaction between cumulative self-motion error and probabilistic learning . If a single learned spatial representation is used for both easterly and westerly runs , it may be hypothesized that re-use of the same map should be evidenced by significant similarity in the response patterns between easterly and westerly runs where the local running directions matched ( Fig 7G ) . However , rat arm-arm correlations with the same local direction ( but opposite global direction ) were weakly correlated ( mean ± SD = 0 . 07 ± 0 . 20 , P = 2 . 4×10−4 , t104 = 3 . 8 , one-sample t-test ) . Surprisingly , probabilistic grid cells show similarly weak correlation ( 0 . 05 to 0 . 15 , 95% C . I . of mean cross-correlation from 104 resamples of 105 probabilistic grid cells ) . The low correlation in probabilistic grid cells shows that re-use of a single map is still compatible with distinct response patterns where the local running direction matched but global direction differed . Taken together , the results suggest that a single laterally compressed spatial map parsimoniously explains the fragmentation of grid maps in hairpin mazes , and is an emergent property of probabilistic learning . Strong visual cues are often used to establish stable grid cell recordings [10 , 11 , 26 , 34–36 , 47] . In rat pups , stable grids emerged after eyelids unsealed and following exploratory experience [51 , 52] . Even tests in darkness were performed in experienced rats [10] , raising the possibility that naïve learning must fail in darkness . However , most experimental arenas have higher-order rotational symmetry in which multiple rotationally-equivalent solutions match sensory information in darkness [18] . This confound is avoided in arenas with 1-fold rotational symmetry ( Fig 8A ) . Thus , SIFM predicts that a naïve rat can learn a spatial representation of a kite-shaped arena in total darkness , and re-localize in darkness when disoriented . Interestingly , resumed learning in light should cause global rotations of grid and boundary fields because novel long-range sensory boundary information cause large discrepancies between predictive and sensed boundary information , destabilizing the distributed estimate of pose to form a new map ( Fig 8A top row ) . A similar mechanism may have contributed to grid remapping during alternating lights on/off training [13] . New probabilistic learning was marked by the emergence of long-range boundary cell responses in light ( Fig 8A bottom row ) . The stability of standard attractor network models depend critically on balanced local connection strengths [8 , 12 , 48 , 53 , 54] , and it is unclear what effects local disruptions of connections may have on the spatial stability of grid cell rate maps . However , disruptions caused by the insertion of tetrodes into grid cell networks do not cause grid field drift [10 , 11 , 13 , 34–36 , 42 , 47 , 51 , 52] , suggesting that underlying computations are robust to local damage . Consistent with rodent results , SIFM tolerates abrupt loss of grid cells either through lesioning 50% of all scale modules , or 50% of grid cells spanning all modules ( Fig 8B ) . Despite stable grids , boundary cell rate maps lost strict adherence to boundary geometry without large-scale grid modules due to increased spatial repetitiveness of the remaining grid code ( Fig 8B , column 1 ) . Grids are predicted to fragment differently in spiral mazes compared to hairpin mazes . In a spiral maze , probabilistic grid maps are uncorrelated along the same running direction ( Fig 8C , 8Is and 8Os ) , distinct from the hairpin maze [36 , 37] ( Figs 6 and 7 ) . Instead , rate maps along arms at the same location but opposite running direction were correlated ( IO ) , encoding place-specific information . Different patterns of grid fragmentation may therefore be signatures of probabilistically-learned spatial information , which vary with environmental structure . Another experiment may be useful for differentiating between SIFM and standard attractor models . Distributed grid codes in SIFM are expected to decouple over time in an unbounded 2D field devoid of visual or other localizing cues . For example , a blindfolded rat which forages in a very large field ( in between contacts with boundaries or other cues ) can only rely on PI to keep track of location . Spatial correlations between SIFM grid cells within a module should decrease gradually due to cumulative PI errors . In contrast , grid cells in standard attractor models should remain functionally coupled and show spatially correlated drift within any one scale module .
Rodent grid cells show consistent and specific properties of probabilistic computations , which include grid fragmentation in hairpin mazes , attractor dynamics , partial grid rescaling in resized arenas , and stable grids in darkness . A new spatial information fusion model ( SIFM ) successfully performed probabilistic learning and recall using grid and boundary cells , unifying diverse grid cell response properties . This contradicts prevailing theories that grid cell networks primarily perform PI , with a separate mechanism correcting cumulative PI errors [9 , 10 , 12 , 14 , 19 , 20 , 22] . The latter implies that a hitherto unidentified spatial system actually solves the hard problem of SLAM . Parsimoniously , SIFM suggests that grid cells can participate directly in SLAM computations to maintain spatial stability . The remarkable similarity between SIFM and rodent data across diverse experiments also show that noisy self-motion [15–18] and boundary vector estimates [27 , 38 , 45] adequately encapsulate the principal sensory information used by rodent grid cells in published experiments . SIFM provides the first demonstration that grid and boundary rate codes suffice to perform probabilistic spatial learning , despite grid codes being surjective functions of position , and boundary inputs varying substantially with the availability of vision . Additionally , probabilistic learning copes with both self-motion and boundary estimation noise , while taking advantage of intrinsic grid phase noise . The breadth and depth of similarity between probabilistic and rodent grid cell responses , partly due to emergent properties of probabilistic learning , suggest that consideration of realistic learning and perceptual constraints can lead to deeper insights into grid cell behaviour and spatial cognition more broadly . A valid question is whether SIFM still performs PI since individual grid cells are based on a PI model which can be considered to be equivalent to a simplified oscillatory interference model ( S1 . 1 . 2 Text ) . A key insight of SIFM is that grid stability depends crucially on maintaining a probability distribution of grid codes which is dynamically modulated through boundary prediction error . Hence PI should be treated as a special case of SIFM in which probabilistic aspects of the computations are disabled ( e . g . , by removing compensatory phase noise in Fig 2 or preventing prediction error feedback in S10 Fig ) , the latter unable to form or maintain stable grids in the presence of sensory noise . It is also important to distinguish between spatial information fusion and simpler PI-reset models . The latter can be considered a special case of information fusion where the probabilistic weight distribution is a Delta function centred on the sensory input . For example , contact with a square arena’s westerly boundary resets the grid code due to previously learned associations between a westerly boundary cell and grid cells [20] . However , such a model rigidly anchors the grid to the boundary , incompatible with partial grid expansion or contraction in resized arenas [34] . Furthermore , PI-reset relies on sensory cues directly driving the correct boundary cell’s activity . In darkness therefore , the drift in grid orientation cannot be slower than the head direction system . Yet in darkness , head direction is unstable within 2 minutes [40] , whereas grid patterns remains stable for 10 to 20 minutes [10] . Finally , associative learning between boundary and grid cells is particularly challenging for any PI-reset model because of the chicken-or-egg problem of SLAM . A noisy PI estimate diverges from true position unless reset via associated boundary cells , but those associations do not exist at the start of learning . This was avoided in a recent model by assuming error-free PI [20] , which is biologically unrealistic . Similarly , place cell based reset information can only be learned if sensory cues are sufficient to drive each place cell’s spatial response [19 , 22] . In darkness , the same difficulties arise , where sensory cues far from a boundary are not sufficient to define position , and grid orientation should drift at least as rapidly as head direction . Hence it is specifically SIFM’s ability to perform probabilistic information fusion , rather than PI or PI-reset , which enables robust learning and recall , which in turn accounts for diverse grid cell response patterns in light and darkness . This work challenges the fundamental assumption of virtually all grid cell models that the computational problem to which grid cells provide a solution is PI . Cues other than self-motion and boundary cues are likely to contribute to spatial stability of cell responses recorded from the hippocampal formation . Arena olfactory cues , for example , contribute to rodent place field stability [23] and may play a role in grid stability , although the latter has not been tested . However , as discussed previously [17] , multiple studies showed that even careful removal of olfactory cues ( including [23] ) did not abolish stable place fields , including in a Morris water maze [24] . In contrast , HD cells drifted even during a single session in darkness where olfactory cues were not specifically minimized [40 , 41] . Taken together , these results suggest that olfactory cues are neither necessary nor sufficient to provide a coherent explanation of stable spatial fields in darkness . A simultaneous recording of grid and HD cells in darkness in a Morris water maze may allow definitive quantification of the contribution of olfactory cues to stable grids in darkness . Since rodent spatial fields are stable across multiple combinations of sensory cues , including without vision or olfactory cues , the underlying computations must be adaptable to variable reliability of each information stream . For example , it would be disastrous for a rat to always learn or recall by relying entirely on either stable visual or olfactory cues since they are not always there . The fact that rodent spatial cognition allows multiple information streams to contribute , and not in an obligatory all-or-none fashion , suggests some sort of probabilistic algorithm . The behaviour of grids under different manipulations and environments provide important tests of any proposed model of rodent spatial learning and recall . In multi-compartment environments , the unexpected emergent boundary representation arising from probabilistic learning resulted in grid fragmentation ( Figs 6 and 7 ) . The learned representation is both a geometric distortion and a form of spatial information compression , in which multiple similar spatial structures are efficiently represented as one . Corridors in an opaque hairpin maze have identical geometry when considering only those walls which are visible to the rat . Thus the sensory boundary information within one corridor is equivalent to a number of other corridors . In the absence of any specific task or contextual disambiguation between the corridors , then arguably there is no need to maintain separate representations of multiple equivalent maze corridors . Potentially less neural resource is required to store a compressed representation . The close match between rodent and probabilistic grid cells in terms of global grid fragmentation , directional dependence , arm-arm-correlation matrices , a variety of arm-arm correlation distributions , and bin-wise 2D rate map similarity patterns shows that this novel and parsimonious explanation must be considered a viable alternative to the original hypothesis of multiple submaps linked at hairpin turns [36] . A working model of the latter has yet to be reported . A large number of attractor network models have been proposed to explain grid cell responses [7–9 , 12 , 20 , 21] . While the functional coupling between grid cells is consistent with attractor dynamics [43] , the mechanism which supports such dynamics remains unclear . Here , biologically-realistic response patterns of probabilistic grid cells rely on functional coupling arising from shared self-motion inputs and feedback from boundary prediction errors , rather than static connections independent of environmental cues . Indeed , the typically rigid synaptic connectivity between attractor network grid cells would lead to near-perfect correlation in grid cell activity , voiding the ability to dynamically track growing uncertainty which is particularly important during prolonged periods in darkness . Furthermore , standard attractor networks require delicately balanced network weights to function so are highly sensitive to local damage , bringing into question their robustness to trauma , disease and even tetrode insertion . Nevertheless , a network implementation of SIFM has not been developed , and it is possible that attractor network properties may be modified to simultaneously enable: functional decoupling to track uncertainty while providing redundancy and robustness; and functional coupling via environmental inputs to maintain stability while explaining diverse grid cell responses . In that way , the shift in an attractor’s activity bump may depend on 1 ) self-motion cues , 2 ) compensatory phase noise which samples phase space , and 3 ) prediction error feedback such as via a boundary code . Intriguingly , cooperative oscillatory and attractor dynamics underpin SIFM function , and may also guide a connectionist instantiation of SIFM . Overall , SIFM shows that a single probabilistic model concurrently and accurately explains numerous grid cell response properties , using realistic noisy inputs , and without assuming prior learned information . Probabilistic spatial computations manifest as a flexible yet stable set of response patterns which depend on arena information and experimental design , often indistinguishable from rodent grid cells . Hence grid cell ensembles may provide a hitherto unexplored window into probabilistic computations in a higher-order cognitive system . The dependence of grid response patterns on sensory inputs also supports the growing view that probabilistic perception complements probabilistic learning [4] . The convergence of experimental and theoretical evidence presented here suggests that spatial perception and spatial learning both depend on probabilistic interactions between grid and boundary cells .
The new spatial information fusion model ( SIFM ) was developed firstly to investigate whether probabilistic fusion of realistic noisy spatial information is possible , even in principle , when constrained by using only representations which can plausibly be encoded by neuronal responses of the hippocampal formation . A second objective of SIFM was to investigate whether rodent grid cell responses are consistent with predictions using probabilistic learning and recall computations , under diverse experimental conditions . The principle of SIFM is fusion of temporally-integrated self-motion information , egocentric boundary vector information , and occasionally head direction information when available , to produce a joint estimate of the current grid code ( pose ) and grid-boundary associations ( map ) distribution ( S1 . 1 Text ) . The specific implementation using noisy sensory cues , neuronal codes and a Rao-Blackwellized particle filter is briefly outlined below ( see S1 . 1 . 2–S1 . 1 . 10 Text for details ) . This is a mathematically succinct implementation of recursive Bayesian inferencing principles , aimed at a systems-level approximation of the computations carried out by neural networks involving grid and boundary cells . Temporally-integrated self-motion information is encoded by a population of grid cells whose responses are modulated by both speed and heading ( S1 . 1 . 2 Text ) . To function , SIFM requires a temporally-stable function of spatial phase , which need not be a regular grid ( e . g . , S4C Fig ) . Noisy self-motion cues provide approximate linear and angular displacement inputs to grid cells , which in turn have independent phase noise which plays a compensatory role ( S1 . 1 . 3 Text ) . Noisy boundary cues provide short-range vectorial information to boundary cells when within somatosensory contact range , and long-range vectorial information when vision is available ( S1 . 1 . 4 Text ) . A noisy compass cue is provided in the presence of vision ( S1 . 1 . 5 Text ) . Grid-boundary associations are approximated by a linear average over time , over a set of predefined map grid codes ( S1 . 1 . 6 Text ) . Each active grid code ( activity of phase-correlated grid cells in multiple modules ) and its association map corresponds to a single ‘particle’ in the particle filter implementation . Using learned grid-boundary associations , new grid codes generate predictive boundary codes , whose discrepancy with sensory boundary information yields a prediction error and importance weight for particle resampling ( S1 . 1 . 7 Text ) . Note that the population of predictive boundary cells have distinct properties from sensory boundary cells which have been described previously [27 , 38 , 45] . Only predictive boundary cell responses are presented . Disorientation is modelled as a random redistribution of grid code activity ( S1 . 1 . 8 Text ) . Pseudocode summarizes key implementation steps for probabilistic learning ( S1 . 1 . 9 Text ) and recall ( S1 . 1 . 10 Text ) . Random simulated trajectories were used to provide full coverage of each arena , mimicking the behaviour of trained rats , except in hairpin mazes where rats were trained to run along maze corridors ( S1 . 2 Text ) . Methods for calculating firing rate maps , spatial crosscorrelograms , gridness index , grid rescaling and border score have been described previously so are only briefly summarized ( S1 . 3 . 1–S1 . 3 . 5 Text ) . The parametric rate map correlation was developed to determine whether a cell’s response is more grid-like or boundary-like , based on computational hypotheses of each cell type ( S1 . 3 . 6 Text ) . Unlike the border score , this metric correctly classified long-range model boundary cells . Associative weight maps were displayed by averaging across all boundary codes at the nominal position corresponding to each grid code of an association map ( S1 . 3 . 7 Text ) . For visual comparison with published data , grid cell and boundary cell spikes were simulated using an inhomogeneous Poisson process ( S1 . 3 . 8 Text ) . Predictive short-range boundary vector maps were produced to visualize the learned local boundary direction ( S1 . 3 . 9 Text ) . Spike-triggered dynamic rate maps and autocorrelograms were used to detect underlying spatial regularity in response patterns which may drift over time ( S1 . 3 . 10 Text ) . Grid phase change was quantified both between grid cells and within the same grid cell across different recall trials in darkness in a kite arena ( S1 . 3 . 11 Text ) .
|
Cells in the mammalian hippocampal formation are thought to be central for spatial learning and stable spatial representations . Of the known spatial cells , grid cells form strikingly regular and stable patterns of activity , even in darkness . Hence , grid cells may provide the universal metric upon which spatial cognition is based . However , a more fundamental problem is how grids themselves may form and stabilise , since sensory information is noisy and can vary tremendously with environmental conditions . Furthermore , the same grid cell can display substantially different yet stable patterns of activity in different environments . Currently , no model explains how vastly different sensory cues can give rise to the diverse but stable grid patterns . Here , a new probabilistic model is proposed which combines information encoded by grid cells and boundary cells . This noise-tolerant model performs robust spatial learning , under a variety of conditions , and produces varied yet stable grid cell response patterns like rodent grid cells . Across numerous experimental manipulations , rodent and probabilistic grid cell responses are similar or even statistically indistinguishable . These results complement a growing body of evidence suggesting that mammalian brains are inherently probabilistic , and suggest for the first time that grid cells may be involved .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"learning",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"brain",
"vertebrates",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"mammals",
"animals",
"rna",
"stem-loop",
"structure",
"cognitive",
"psychology",
"sensory",
"physiology",
"rna",
"structure",
"hippocampal",
"formation",
"molecular",
"biology",
"biochemistry",
"rna",
"psychology",
"rodents",
"sensory",
"cues",
"macromolecular",
"structure",
"analysis",
"cell",
"biology",
"anatomy",
"nucleic",
"acids",
"physiology",
"biology",
"and",
"life",
"sciences",
"sensory",
"perception",
"cognitive",
"science",
"amniotes",
"organisms",
"cell",
"fusion"
] |
2016
|
Probabilistic Learning by Rodent Grid Cells
|
The nuclear pore complex ( NPC ) provides the sole aqueous conduit for macromolecular exchange between the nucleus and the cytoplasm of cells . Its diffusion conduit contains a size-selective gate formed by a family of NPC proteins that feature large , natively unfolded domains with phenylalanine–glycine repeats ( FG domains ) . These domains of nucleoporins play key roles in establishing the NPC permeability barrier , but little is known about their dynamic structure . Here we used molecular modeling and biophysical techniques to characterize the dynamic ensemble of structures of a representative FG domain from the yeast nucleoporin Nup116 . The results showed that its FG motifs function as intramolecular cohesion elements that impart order to the FG domain and compact its ensemble of structures into native premolten globular configurations . At the NPC , the FG motifs of nucleoporins may exert this cohesive effect intermolecularly as well as intramolecularly to form a malleable yet cohesive quaternary structure composed of highly flexible polypeptide chains . Dynamic shifts in the equilibrium or competition between intra- and intermolecular FG motif interactions could facilitate the rapid and reversible structural transitions at the NPC conduit needed to accommodate passing karyopherin–cargo complexes of various shapes and sizes while simultaneously maintaining a size-selective gate against protein diffusion .
The nuclear pore complex is a supramolecular protein structure in the nuclear envelope that controls nucleo-cytoplasmic traffic and communication ( Figure 1A ) [1] . A key NPC architectural feature is a poorly understood semi-permeable diffusion barrier at its center , which allows passive diffusion of particles less than 3–4 nm in diameter ( or 30–40 kDa in mass for a folded protein ) and opens to allow facilitated transport of larger particles up to 39 nm in diameter [2] . The NPC is composed of ∼30 proteins or nucleoporins ( nups ) that are present in multiple copies [3] , [4] . Among these , a group that contains numerous phenylalanine-glycine repeats ( FG nups ) ( a subset is shown in Figure 1B ) line the transport conduit of the NPC ( Figure 1A ) . These FG nups function as stepping-stones for karyopherin movement across the NPC [5] , [6] and as structural elements of the NPC protein diffusion barrier [7] , [8] . The three dimensional structure of S . cerevisiae FG nups is unusual because their 150–700 amino acid ( AA ) FG domains are natively unfolded [9] in their functional state [6] . Since there are ∼150 FG nups in each NPC [4] , it is currently hypothesized that its transport conduit is lined and/or flanked by 150 natively unfolded FG domains . Together these FG domains constitute ∼12% of the total NPC mass or >6 . 5 MDa of its ∼55 MDa structure in yeast [10] . The FG domains of nups were initially hypothesized to function as repulsive entropic bristles that create a virtual gate at the NPC periphery [11] , [12] , and later as cohesive polypeptide chains that form a hydrogel at the NPC center [8] , [13] , [14] . More recently , an analysis of all nup FG domains in S . cerevisiae indicated that some FG domains ( the GLFG-rich domains ) bind to each other weakly via hydrophobic attractions between their FG motifs , whereas other FG domains ( the FxFG-rich domains ) do not form such cohesions [7] . Despite the fact that different subtypes of FG domains are defined by their content of FxFG , GLFG or SAFGxPSFG motifs , their ability to interact with each other ( i . e . , their cohesiveness ) seems to correlate best with the AA composition of the sequences between FG motifs , rather than with the specific FG motif [7] . Hence , the human FG nups may also interact with each other , despite having only one GLFG-rich nup among its eleven members [3] . It is generally assumed that natively unfolded proteins have some preferred 3-D structures dictated by intra-molecular cohesion [15] , [16] . Current evidence that the FG domains of nups have some structure is based on CD and FTIR spectroscopic analysis , which indicates that FG domains have anywhere from 5% to 20% α-helical and β-sheet content at any given moment [9] , yet the locations of such structures in the protein are probably ever-changing . The conformational flexibility inherent to natively unfolded proteins and protein domains such as those in the FG nups , places them beyond the reach of classical structural biology tools such as X-ray crystallography and homology-based computational methods [17]–[20] . However , it is clear that these and other unfolded proteins participate in a wide range of key cell biological processes [21]–[23] and that their native plasticity bestows specific functional properties , such as rapid molecular interaction times and the ability to bind multiple proteins simultaneously [24] . In the case of nucleoporin FG domains , a key function is to bind multiple karyopherins [6] with very rapid interaction times [25] . Thus , in contrast to folded proteins , the structure of natively unfolded proteins must be described as a dynamic ensemble of interconverting conformers . Since traditional experimental methods for elucidating protein structure cannot be used with natively-unfolded proteins , new approaches are needed to study and describe their dynamic ensemble of structures . In this emerging area of research , Jha et al [26] , [27] have recently introduced a general statistical coil model , and Bernado et al [28] , [29] have estimated the nuclear magnetic resonance ( NMR ) measured residual dipolar couplings ( RDCs ) [30] from dynamic simulations to characterize the ensemble-averaged conformations of α-synuclein . Also , Ollerenshaw et al [31] have applied a native-centric topological model to understand the essential folding/unfolding dynamics SH3 domains , and Pappu and co-workers have characterized poly-glutamines as a function of chain-length conformational sampling by molecular dynamics ( MD ) and Monte Carlo simulations [32] . Most of these computational investigations suggest the existence of a preferred ensemble of conformers for each protein , rather than suggesting pure random coils . Here we conducted molecular dynamics simulations and biophysical measurements on a small FG domain from the yeast nucleoporin Nup116 ( Q02630 ) to test the hypothesis that phenylalanines in its FG motifs function as intramolecular cohesion elements that impart structure . Apart from its cell biological significance , we chose this protein as a model system to investigate how a combination of molecular dynamics simulations and biophysical measurements can be used to characterize the ensemble of structures adopted by a natively unfolded protein , such as the FG domain of a nucleoporin .
Twenty independent MD simulations were performed at 300 K ( 25°C ) on the wild-type ( 6 ns ) and F>A mutant ( 5 ns ) versions of a Nup116 FG domain ( AA 348–458 ) starting from a fully-extended conformation . The goal of these simulations was to sample the conformational distribution of the proteins as close as possible to their native distribution in solution . As soon as the simulations started , within the first 100 ps , the extended FG domains collapsed into a more cohesive or compact ensemble of structures with small patches of unstable ( see below ) secondary structure . Since the wild-type and mutant FG domains are highly flexible and disordered , the resulting end-structures from each of the twenty simulations did not resemble one another as expected for natively unfolded proteins ( see Figure 1C for representative examples ) . Despite the fact that the nup structures were ever-changing ( see below ) , the ensemble of structures for each did “converge” to a similar size early in the simulation according to various metrics of size , which changed little in the last 3 ns . This was evidenced by a constant radius of gyration ( Figure S1 ) and by statistical analyses that showed no significant change in the range of Rg values during the last 3 ns ( data not shown ) . To describe quantitatively the structural dynamics of the FG domains , we calculated the auto-correlation function of a vector of the 118 Φ and 118 Ψ angles along the peptide backbone of FG domain structures sampled every 1 ps from the MD trajectory . Figure S2 shows the autocorrelation functions with a 200 ps window from the final 3 ns of simulation of all twenty wild-type and F>A mutant FG domain simulations , along with the comparable auto-correlation function from the MD simulation of a control protein that is folded ( fibroblast growth factor 1 ) . For each of the replicate nup simulations , the correlation in the Φ–Ψ angles dropped from 1 to 0 . 738 ( ±0 . 031 ) or 0 . 741 ( ±0 . 023 ) in 1 ps for the wild-type or mutant FG domains , respectively , and then slowly decayed over 200 ps to 0 . 672 ( ±0 . 037 ) or 0 . 665 ( ±0 . 029 ) , respectively . In contrast , for the control protein fibroblast growth factor 1 , the autocorrelation function dropped to only 0 . 875 in 1 ps , and then to 0 . 864 over the 200 ps auto-correlation window . These results indicate that the AA chain backbone of wild-type and mutant FG domains is constantly changing structure and is highly dynamic in comparison to a folded protein . The ensemble of structures for each of the twenty MD trajectories generated for the wild-type and mutant FG domains were sampled at 1 ps intervals during the final 3 ns of the simulations , yielding a total of 60 , 000 structures for each protein . The secondary structure content was then analyzed in detail to determine the fraction of time during the simulations that each AA residue spent as part of a “helical” structure ( either an α-helix or a 310-helix ) . In general , no significant difference in overall helical content between wild-type and mutant FG domains was observed . The alpha- and 310- helical structures that did form ranged in size from 2–6 AA residues and did not persist for more than 35 ps on average ( data not shown ) . The maximum duration of an α-helix and a 310-helix was 97 and 699 ps , respectively ( data not shown ) . Using the same set of 60 , 000 structures , two measures of protein compactness were calculated: the radius of gyration ( Rg ) and the end-to-end distance between terminal residues . The average ( ±1 standard deviation ) end-to-end distance for the wild-type FG domain simulated at 300 K was 20 . 42 Å ( ±9 . 51 ) , and for the mutant was 20 . 69 Å ( ±7 . 78 ) ( data not shown ) . The predicted radius of gyration was 14 . 52 Å ( ±1 . 18 ) for the wild-type and 14 . 41 Å ( ±1 . 24 ) for the mutant FG domain ( Figure 2A ) . The simulations sampled different regions of conformation space because significant run-to-run variations were observed in the probability distributions for each structural parameter . The similar Rg and end-to-end distance values obtained for the wild-type and mutant FG domains implied that both proteins occupy equivalent hydrodynamic volumes . However , this conclusion was at odds with two different quantitative measurements of the physical dimensions of purified FG domains ( see below ) . Interestingly , it has been reported that increasing MD simulation temperature can yield protein dimensions that more closely resemble those obtained by NMR protein conformation measurements [33] , [34] . Indeed , when we extended the nup MD simulations for an additional 1 ns at 325 K ( 52°C ) or at 350 K ( 77°C ) , a very different picture emerged ( Figure 2A ) . At 325 K there was a slightly greater difference in the average radius of gyration between the wild-type ( 15 . 11±1 . 43 Å ) and the mutant ( 15 . 76±2 . 58 Å ) FG domains . Five of the twenty mutant simulations now had an average Rg greater than 18 Å , but all the wild-type simulations had an average Rg below 18 Å , indicating that the mutant FG domain is larger ( Figure 2A ) . In addition , the average end-to-end distance for the wild-type FG domain was 20 . 84 Å ( ±10 . 75 ) compared to 24 . 26 Å ( ±13 . 16 ) for the mutant ( data not shown ) . At 350 K , there was a much greater difference between their radii of gyration ( Rg ) . The average Rg was 17 . 40 Å ( ±3 . 11 ) for the wild-type FG domain and 23 . 68 Å ( ±6 . 05 ) for the mutant domain ( Figure 2A and 2B ) . At 350 K , fifteen of the twenty mutant simulations had an average Rg greater than 20 Å , compared to only three for the wild-type simulations ( data not shown ) . Consistently , the average end-to-end distance for the wild-type FG domain was 29 . 95 Å ( ±16 . 04 ) compared to 52 . 56 Å ( ±25 . 31 ) for the mutant ( Figure 2C ) . The larger dimensions obtained for the wild-type and mutant FG domains at 325 K and 350 K compared to 300 K were likely due to thermal “melting” during the additional 1 ns of simulation . These data combined provide a first indication that the F>A mutant Nup116 FG domain is not as intramolecularly cohesive or compact as the wild-type version . As a way of assessing the dynamic structure of the FG domain , particularly from the point of view of the FG motifs , we plotted the distances between the backbone β-carbons ( Cβ ) for the ten sites that correspond to the phenylalanine ( Phe , F ) or to the substitute alanine ( Ala , A ) residues in the various FG motifs . The distances used were from the MD simulations at 350 K , which yielded structures ( Figure 1C ) that better reproduced the dimensional difference between the wild-type and mutant FG domains measured by NMR analysis and in sieving columns ( see below ) . The distance analysis yielded 45 F–F ( or A–A ) distances for each structure ( Table S1 ) . Probability distributions for each Cβ-to-Cβ distance were calculated and analyzed looking for significant differences between the wild-type and mutant FG domain configurations . To estimate the sharpness of the Cβ-to-Cβ distance distributions , the number of 1 Å wide bins that had greater than 10% of the probability distribution was counted; no bin had more that 20% of the probability distribution . This metric was calculated for all 45 Phe–Phe Cβ-to-Cβ distances in all twenty replicates of the FG domain simulations . In stable tertiary structures , these distances occur as one sharp-peak distribution around the equilibrium inter-residue distance; in a fully random ensemble of structures , they occur as a very broad distribution; and in semi-structured proteins , they occur as one or more intermediate-width distributions . Figure 3A shows two representative examples of the probability distributions obtained from the MD trajectories at 350 K; the values shown correspond to the distribution of distances between phenylalanine F84 and F93 in the wild-type FG domain ( Figure 1C ) or alanine A84 and A93 in the F>A mutant domain . Overall , for the wild-type FG domain simulations , the majority of the simulations analyzed had more than three peaks; by comparison , only a minority of the F>A mutant simulations analyzed exhibited a similar level of sharpness in the distance distributions ( data not shown ) . These results provide tentative evidence that the wild-type FG domain is more structured than the mutant . Repeating this analysis to include only peaks with the distance distributions of <15 Å yielded a very similar result ( data not shown ) . To permit comparisons of the average inter-residue distances , the probability distributions obtained for the Cβ-to-Cβ distances were fit to a single Gaussian distribution even though in some cases there were multiple distinct peaks ( Figure 3A ) . This was only a rough approximation to the observed probability distribution , but the assumption was justified in the context that these ensemble of structures were to be used ( see below ) . After all , these structures are rapidly inter-converting and the width of the Gaussian is broad enough to accommodate all of the major peaks in the distribution . For example , in the case of the F84–F93 distance distribution , a Gaussian centered at 16 . 2 Å with a width of 11 Å covered both peaks at 10 and 20 Å ( Figure 3A ) . Probability distributions of all inter-residue distances obtained from the MD simulations were subjected to clustering using the Pearson squared correlation . This was done to determine how any two of the distributions sampled in regular intervals of the MD simulations are correlated with each other . The correlation coefficient does not depend on the specific measurement units used because other correlation coefficients , such as Euclidian distance metric , yielded similar clustering effects ( data not shown ) . Figure 3B shows the matrix intensity plots of the correlations for the wild-type and mutant FG domains . The indices of the matrix correspond to the various Phe–Phe pairs ( or Ala–Ala pairs ) listed in Table S1 . Indices 1 through 9 correspond to the distance from the first Phe ( at position 13; see Figure 1C ) to the other 9 Phe ( positions 23 through 113 ) , while indices 10–18 correspond to similar distances from the second Phe ( at position 23 ) to the other eight Phe's ( positions 32 through 113 ) and so on . Altogether , the clustering analysis showed that there is a stronger correlation between the various Phe–Phe distributions in the ensemble of wild-type FG domain structures than between the various Ala–Ala distributions in the ensemble of mutant FG domain structures . This indicated that the wild-type FG domain is generally more ordered than the mutant . To obtain a broader view of the dynamical correlation between inter-residue distances in the FG domains , the Pearson correlation coefficient between all 990 distinct interresidue distances in all 20 simulation replicates of each FG domain were calculated , yielding 19 , 800 correlation coefficients . In this case , a value of 1 . 0 would indicate a perfect linear correlation between two inter-residue distances ( as one inter-residue distance grew larger , the other would grow by a proportionate amount ) ; a value of 0 . 0 would indicate no correlation between a distance pair; and a value of −1 . 0 would indicate perfect anticorrelation . Figure S3 shows back-to-back histograms of the resulting correlation coefficients . For the wild-type FG domain , 9 . 0% of the observed correlation coefficients had values above 0 . 7 versus 7 . 1% in the F>A mutant . This demonstrated that the ensemble of wild-type FG domain structures shows a bias towards higher correlation coefficients between inter-residue distances than the F>A mutant domain , indicating more structural coherence in the wild-type FG domain than in the mutant . To better describe the relationship between FG motifs in the FG domain , the distances between F–F pairs ( or substitute A–A pairs ) were also categorized into groups representing distances of 10–15 , 15–20 , or >20 Å . These are shown in Figure 3C as thick red , medium blue , or thin green lines , respectively . A list of all distances for both proteins is given in Table S2 . Among the F–F distances in the wild-type FG domain , seven F–F pairs are less than 15 Å apart ( red text in Table S2 and red-thick lines in Figure 3C ) . In contrast , the F>A mutant FG domain had only two A–A pairs with such short distances . In the wild-type FG domain , four F–F pairs were in the range of 15–20 Å apart ( blue ) , while seven F–F pairs were farther than 20 Å ( green ) . In the F>A mutant , there were eight A–A pairs with distances in the mid-range ( 15–20 Å ) , and three pairs showing distances greater than 20 Å . These results demonstrate that the intramolecular distances between FG motifs in the wild-type and mutant FG domains are quantifiably different from each other . There was a tendency for FG motifs in the wild-type FG domain to be proximal to each other ( i . e . , to cluster ) , which was absent in the mutant . This conclusion is consistent with the hypothesis that the FG motifs in the wild-type Nup116 FG domain interact intra-molecularly in a manner similar to what has been observed for intermolecular interactions between this Nup116 FG domain and other FG domains of nups [7] . The structural predictions made by the in silico modeling prompted us to seek physical evidence that the phenylalanine residues in FG motifs function as structural cohesion elements that form putative intra-molecular interactions within the Nup116 FG domain . In principle , a change in the dynamic ensemble of FG domain structures resulting from the substitution of all Phe's to Ala's could be detected by NMR . A less-ordered mutant FG domain would exhibit a slower diffusion coefficient . The wild-type and F>A mutant versions of the Nup116 FG domain were purified to `homogeneity and subjected to NMR analysis . Plots of the one-dimensional 1H NMR spectra are shown in Figure 4 ( left panels ) . It was anticipated that the hydration of the FG domains would be significantly different from that of ordered , globular proteins [35] , [36] due to the lack of stable folded structures in the FG domains [9] . When presaturation of the water was used there was a significant reduction in the intensity of the amide region of the nup spectrum due to fast exchange with the solvent protons [37] , [38] . This observation , combined with the narrow chemical shift dispersion of the amide resonances ( 7 . 9–8 . 5 ppm ) in both nup spectra , was a clear indication that both FG domains are natively unfolded and highly dynamic . NMR experiments conducted at lower temperatures ( 5 and 10°C , compared to 25°C ) gave similar results , but offered no significant improvement in the spectral dispersion ( data not shown ) . Experimental self-diffusion measurements ( intensity vs . product of the area of gradient pulse strength and the diffusion length ) of the FG domains and the corresponding exponential fits are also shown in Figure 4 ( right panels ) . The data yielded self-diffusion coefficients ( Dsexpt ) values of 13 . 17 ( ±0 . 26 ) and 12 . 18 ( ±0 . 12 ) ×10−11 m2 s−1 for the wild-type and mutant FG domains , respectively . This indicated slower diffusion for the less-ordered mutant FG domain . Despite the mass of the F>A mutant domain being smaller ( 11 . 9 kDa ) than wild-type ( 12 . 6 kDa ) ( due to the replacement of 10 Phe for Ala ) the diffusion constant of the mutant was smaller on average , suggesting that its effective hydrodynamic volume is larger . As expected for unfolded proteins [39] , the diffusion of the wild-type and mutant FG domains was significantly slower than a folded protein of higher molecular weight [39] , [40] , indicating that the wild-type and mutant FG domains have unfolded structures that sample a relatively large conformational space . To further characterize the hydrodynamic properties of the wild-type and mutant Nup116 FG domains , each was analyzed by FPLC in a sieving column to determine its Stokes radius . The expectation was that the less ordered mutant FG domain would occupy more hydrodynamic space and would elute faster from the sizing column . Purified wild-type and mutant versions of the Nup116 FG domain were subjected to size-fractionation through an FPLC Superdex 75 column and their elution profiles were compared to that of commonly-used size standards , such as carbonic anhydrase ( 29 kDa , Rs = 23 . 5Å ) , ovalbumin ( 45 kDa , Rs = 29 . 8Å ) , and BSA ( 68 kDa , Rs = 35 . 6Å ) . The Stokes radius for the wild-type FG domain was measured at 25 . 2 ( ±0 . 6 ) Å ( Table 1 ) , which is larger than carbonic anhydrase despite the FG domain having less than one-half the mass . This highlighted the fact that the FG domain is natively unfolded . The F>A mutant domain eluted faster from the sieving column and migrated as a particle with a Stokes radius equivalent to 27 . 1 ( ±0 . 6 ) Å , which is larger than the wild-type FG domain despite the mutant having less mass ( Table 1 ) . This apparent loss of compaction for the mutant FG domain compared to the wild-type ( a ∼20% change in hydrodynamic volume ) was consistent with its slower NMR diffusion coefficient ( Figure 4 ) and with the computationally-predicted difference in hydrodynamic dimensions between them at 350 K ( Figure 2 ) . The observed loss of intra-molecular cohesion in the mutant FG domain supported our hypothesis that FG motifs within the natively-unfolded FG domain of Nup116 interact intramolecularly via phenylalanines that cluster through hydrophobic attractions . In essence , the hydrophobic interactions between FG motifs likely bias the arrangement of coils within an FG domain to form an ensemble of dynamic non-random tertiary structures with a quantifiable level of intramolecular cohesion . The hydrodynamic volume or Stokes radius of a protein in different structural configurations ( e . g . , a folded globule , a molten globule , a premolten globule , a coil , an extended coil ) can be estimated from its mass using mathematical equations [41] . These equations were derived from the analysis of large data sets containing experimentally-determined hydrodynamic values for proteins in those structural configurations . Here , using the mass of the wild-type and mutant Nup116 FG domains , we calculated their hypothetical Stokes radius in each structural configuration and compared these predicted values to our experimentally-measured Stokes radii values ( Table 1 ) . The goal was to identify the structural configuration of each FG domain that best matched the biophysical measurement obtained for its hydrodynamic volume . For the wild-type FG domain , a predicted native pre-molten globule structure matched best its measured Stokes radius , and for the F>A mutant , a predicted native coil structure was the best match ( gray boxes , Table 1 ) . These results support the hypothesis that GLFG motifs in nucleoporins function as intra-molecular cohesion elements , because their absence caused a loss of compaction in the Nup116 FG domain , shifting its dynamic ensemble of structures from native premolten globular configurations to native coil configurations .
We have used a combined computational and biophysical approach to characterize the dynamic ensemble of structures adopted by a natively unfolded or intrinsically unstructured protein . Specifically , we characterized the ensemble of conformations adopted by a fragment of the FG domain of the S . cerevisiae nucleoporin Nup116 ( AA 348–458 ) and of a mutant version thereof ( F>A ) lacking the phenylalanines in its predominantly GLFG motifs ( Figure 1 ) . Both FG domains were found to be highly dynamic and disordered , yet contained quantifiable structural differences between them . The MD simulations predicted a more cohesive and/or compact ensemble of structures for the wild-type FG domain compared to the F>A mutant based on the average radius of gyration and end-to-end distances ( Figure 2 ) . This structural prediction was supported by the inter-phenylalanine or inter-alanine distance analysis ( i . e . , the distance between wild-type or mutant FG motifs , respectively ) ( Figure 3C ) , which indicated shorter distances between the FG motifs in the wild-type domain; and by the Pearson correlations of F–F ( or A–A ) pair distances in the FG domains ( Figure 3B ) , which indicated that the Nup116 FG domain has increased probability of sampling geometries that are more ordered when the phenylalanines in the FG motifs are present . The structural predictions made by the MD simulations were confirmed by direct physical examination of purified FG domains , through NMR-based measurement of their hydrodynamic properties ( Figure 4 ) and by measurement of their hydrodynamic radii in sieving columns ( Table 1 ) . In all of the analyses , the wild-type FG domain was found to be more compact than the mutant domain . Unlike the simulations at 350 K , the lower temperature simulations ( e . g . , at 300 K ) did not reproduce this difference in hydrodynamic volumes . Hence , the MD simulations at the higher temperature ( 350 K ) for this class of natively-unfolded proteins may reproduce more accurately their physical properties in solution . As a caveat , the magnitude of the size difference between the nups simulated at 350 K is larger than the magnitude of the difference in their physical dimensions as measured in the sizing columns ( Table 1 ) . Notwithstanding , the simulation values matched , within the experimental error of the simulations ( s . d . ±18% ) , the measured Stokes radii for the purified FG domains . Since only 20 single-molecule simulations were used to predict the dimensions of the FG domains , whereas ∼35 trillion molecules were used to accurately measure their average dimension in the sieving columns ( s . d . ±2% ) , it seems likely that a greater number of simulations for greater time-periods could increase the congruency between simulated and measured values . The mass and physical dimensions of the Nup116 FG domain fragment analyzed here ( AA 348–458; MW = 12 . 6 kDa; Rs = 25 . 2±0 . 6 ) , together with the scaling relations developed by Uversky's group [41] , led us to conclude that this Nup116 FG domain fragment is best described as a dynamic ensemble of native pre-molten globular structures ( Table 1 ) . This structural information can in turn be used to predict the physical dimensions of the full-length Nup116 FG domain ( AA 1–960; see [42] ) based on its mass and assuming that it also adopts native premolten globular structures . Using the scaling relations , which convert protein mass to physical dimensions in any of a number of structural configurations [41] , we estimated that the entire Nup116 FG domain would occupy a hydrodynamic volume equivalent to a 12-nm-diameter sphere ( Table S3 ) . For comparison , its volume would be equivalent to a 16-nm-diameter sphere if it were to adopt less compact native-coil configurations; or to a 19-nm-diameter sphere if it were to adopt extended-coil configurations; or to a 7-nm-diameter sphere if it adopted a tightly folded configuration ( data not shown ) . Likewise , size estimations can be done for other full-length nucleoporin FG domains that have similar AA composition and FG motif type as Nup116 ( e . g . , the GLFG nup subfamily shown in Figure 1B ) [42] . Such analysis predicts that their FG domain dimensions would be equivalent to spheres with diameters of 7 , 7 , 11 , and 7 nm for Nup49 ( AA 1–251 ) ( Q02199 ) , Nup57 ( AA 1–255 ) ( P48837 ) , Nup100 ( AA 1–800 ) ( Q02629 ) , and Nup145n ( 1–216 ) ( P49687 ) , respectively , assuming native premolten globular configurations for each case ( Table S3 ) . These predicted dimensions for the FG domains are generally consistent with the 16–46% larger dimensions reported for the full-length FG nups containing the FG domain , the folded NPC anchoring domain and a Protein A tag ( Table S3 ) [43] . Interestingly , all of these FG domains including the Nup116 FG domain appear to be large enough to butt against each other locally within the NPC ( at least within a single spoke and probably between adjacent spokes ) given their close anchoring at the NPC ( see paragraph below ) [43] , [44] , yet appear to be too small to span across the NPC transport conduit from their tether sites within the NPC scaffold ( ∼19 to 32 nm away from the conduit center; Table S3 ) to the space occupied by the FG domains of nups anchored at the opposite side ( see Figure 1A and Figure S5A ) . This is because the NPC transport conduit has an estimated radius of 19 nm [2] , [43] , [44] , which is significantly larger than the estimated diameter for these FG domains ( ∼7–12 nm ) . Notwithstanding , large fluctuations in the dimensions of FG domains , which are intrinsic to natively unfolded structures , or steric hindrance effects caused by the spatial confinement between closely-anchored FG domains [12] , [45] could cause the FG domains to extend further out into the transport conduit ( Figure S5B ) . The cohesive properties between FG domains within the conduit [7] , [14] ( also see below ) , or the cross-linking action of karyopherins within the conduit ( i . e . , karyopherins appear to bind multiple FG motifs in different FG domains simultaneously ) [45]–[47] could transiently stabilize some of the extended FG domain conformations ( Figure S5B and S5C , respectively ) [14] . The evidence presented here suggests that the FG motifs in Nup116 function as structural , intramolecular cohesion elements that bias the arrangement of coils within the FG domain and condense its dynamic ensemble of structures into more cohesive , less disordered states . In the case of the Nup116 FG domain examined , its FG motifs are responsible for shifting its ensemble of structures from native-coil configurations ( as seen for the mutant ) to native pre-molten globular configurations ( Table 1 and Figure 1C ) . In principle , all types of FG motifs ( GLFG , FxFG , SAFG , PSFG , etc . ) [42] could exert cohesion through hydrophobic pairing , stacking , zippering , or otherwise clustering of the aromatic ring of phenylalanine side chains through energetically favorable aromatic edge-to-face interactions , as opposed to less favorable face-to-face ( π–π ) interactions [48] . Interestingly , a report by Dhe-Paganon et al . , defined a “phenylalanine zipper” motif within the hydrophobic core of APS , which is critical for APS dimerization [49] . There , the aromatic side chains of ten phenylalanine residues are uniquely stacked to form a zipper that is stabilized by helical secondary structures in the protein backbone . Although FG domains do not appear to have stable secondary structures , residues surrounding the FG motif , such as the leucine residue of GLFG motifs or the second phenylalanine residue in FxFG motifs , could enhance the hydrophobic clustering effect by increasing the local hydrophobicity of the Phe residue in the FG motif and/ or by influencing the orientation of its Phe ring . A two dimensional representation of the Nup116 FG domain AA sequences in a hydrophobic cluster analysis ( HCA ) [50] , [51] illustrates its hydrophobic “LF” patches very well ( Figure S4 ) . Although HCA is most commonly used in determining hydrophobic clusters in helical patterns [31] , it is also informative in the absence of a structural fold because it allows the identification of hydrophobic features between nearby AAs . The HCA analysis highlighted LF patches and MFMF-connections in the wild-type Nup116 FG domain , which were missing in the F>A mutant domain . This implied that the F>A mutant FG domain is less compact because it does not have hydrophobic patches and connections to make intra-molecular interactions . Our finding that the FG motifs can function as intramolecular cohesion elements has important implications for the general architecture and function of the NPC , especially if the ability of these FG motifs to mediate intramolecular cohesion of coils functionally mimics their demonstrated ability to mediate intermolecular cohesion between FG domains [7] . Indeed , the representative FG domain of Nup116 analyzed here and the FG domains of other ( but not all ) FG nups ( Nup49 , Nup57 , Nup100 , nNup145 , and Nup42 ) engage in homotypic and heterotypic interactions with each other in vitro and in vivo via FG motifs [7] . By analogy to the Nup116 FG domain , this group of cohesive FG domains may also exhibit intramolecular cohesion of their own FG motifs . At the NPC , such intra- and intermolecular FG motif interactions could be in competition with each other , possibly causing the FG domains to fluctuate between monomeric and polymeric states ( Figure S5B ) . Alternatively , such interactions could be in a dynamic equilibrium with each other to form a metastable quaternary structure ( Figure S5A ) . According to the two-gate and the hydrogel models of NPC architecture , the FG motifs of nups within the NPC conduit engage in intermolecular cohesions with each other to form a highly flexible network of cohesive polypeptide chains , which forms a size-selective sieve or gate [7]–[9] , [13] , [14] . The cohesiveness of such network ( s ) is presumably maintained by the weak but numerous interactions between FG motifs [52] . However , if the intra- and intermolecular interactions between FG motifs were in competition with each other at the NPC , then intramolecular cohesions could effectively prevent the FG domains from forming a network altogether by causing them to “fold back” on themselves ( i . e . , an autoinhibitory mechanism ) . What type of FG motif interaction dominates at the NPC , either intramolecular or intermolecular , is indeed an important question whose answer may rely largely on four parameters: the distance between FG domain anchor sites at the NPC , the volume of space occupied by each FG domain , the space available at the NPC for each FG domain , and the steric hindrance effect between neighboring FG domains [12] , [45] , [53] . As discussed above , the estimated dimensions for the “cohesive” GLFG-rich domains of yeast nups ( 7–12 nm diameter spheres ) combined with the close proximity between their anchor points within each spoke ( most are ≤5 nm apart from others anchored adjacently and ≤10 nm from others anchored above or below in the z-axis; see Table S3 ) implies that at least within a spoke and probably between adjacent spokes the FG domains of these nups butt against each other to occupy overlapping space [43] , [44] . This close positioning could allow or even promote the formation of a supra-molecular quaternary structure of cohesive FG domains at the NPC through a multitude of inter-molecular FG motif interactions . This structural assembly could take the form of a meshwork of intertwined polypeptide chains [14] , [52] , or alternatively , based on the data presented here , the assembly could take the form of a doughnut-shaped array of laterally-cohesive , native pre-molten globules ( as depicted in Figure 1A and Figure S5A ) . Most importantly , local reversible shifts in the equilibrium between intra- and intermolecular FG motif interactions could facilitate the fast structural changes in the NPC permeability barrier , which are presumably coupled to the passage of karyopherin-cargo complexes of different shapes and sizes during transit across the NPC ( Figure S5C ) . As karyopherin–cargo complexes disrupt ( either by mass action or by direct interaction with the nup FG domain ) intermolecular FG motif interactions during transit ( as predicted for all FG nups in the hydrogel model , or for a discrete subset of nups in the two-gate model ) [7] , [8] , [14] , the FG motifs liberated as a result would become available to form intramolecular interactions . This could cause the FG domains to fold back on themselves ( i . e . , to compact ) , effectively opening the permeability barrier by suddenly occupying less space . It remains to be determined whether other types of nup FG domains , which do not display intermolecular cohesions with each other via FG motifs [7] , can nevertheless form intramolecular cohesions of their own FG motifs to adopt compact configurations . According to the “virtual gate” [11] , the “oily spaghetti” [54] and the “two gate” [7] , [55] models of FG domain architecture , these FG domains would exist as highly-extended polypeptide chains , as observed for the Xenopus Nup153 FG domain [12] . Interestingly , in the case of Nup153 , its extended FG domain appears to compact upon binding a karyopherin [45] . Clearly , a more detailed structural characterization of the various FG domains as they interact with karyopherins and each other is needed to fully understand the dynamic and highly flexible structure of the NPC transport conduit .
MD simulations of individual FG domains were started from a fully extended backbone structure ( i . e . , with the Φ and Ψ angles set to 180° for all residues except for the three proline residues , which put a 60° bend in the sequence ) . A different random number seed was chosen for each of the different simulations to randomize the initial atom velocities . Twenty separate simulations were run for either 6 ns ( wild-type ) or 5 ns ( mutant ) each using different initial atomic velocities and analyzed at 1 ps intervals . The wild-type fragment required an additional nanosecond of dynamics to have its radius of gyration converge . All MD simulations were performed with AMBER [56]–[58] using implicit solvent models . Each molecule was simulated in the presence of a Generalized Born/Surface Area ( GB/SA ) implicit solvent model [59] that calculates an effective solvation energy as an empirical parameter multiplied by the exposed surface area of different atom types . Each molecule was simulated using the GB/SA implicit solvent implementation in Amber versions 7 and 8 . Each system is energy-minimized using 100 cycles of conjugate gradients . Constant-temperature molecular dynamics at 300 K with a coupling constant of 2 . 0 ps was performed on the minimized systems using the standard partial charges for the Amber force field and Bondi radii for the atoms . Bonds containing hydrogens were constrained using SHAKE and a time step of 2 fs was used in all simulations . A cutoff of 250 Å was used for the electrostatic interactions , which for this system is equivalent to infinity . The salt concentration ( Debye-Huckel screening ) was set at 0 . 15 M . Secondary structure analysis: For the final 3 ns of each simulation , the structure was analyzed every 1 ps using a standard program for identifying secondary structure from atomic coordinates ( Define Secondary Structure of Proteins; DSSP ) [60] . Radius of gyration and end-to-end distance analyses: For the final 3 ns of each simulation , radii of gyration and the end-to-end distances between terminal residues were calculated using the program CARNAL and ptraj , distributed with AMBER 7 [57] , [58] . High temperature molecular dynamics simulations: Molecular dynamics were performed at elevated temperatures for each of the 20 wild-type and mutant FG domain simulations . The GB/SA simulations were all restarted after 5 ns coupled to a heat bath at 325 or 350 K with all other parameters of the simulation kept the same . The simulations were run for 1 ns , and the final 500 ps were used for analysis . To determine the degree of dynamical change in the ensemble of FG domain structures , the autocorrelation function was derived for a vector ( total vector length = 236 ) composed of the 118 Φ and 118 Ψ angles ( in the 111 AA nucleoporin sequence with a 9 AA N-terminal tag; see Figure 1 ) along the peptide backbone of structures sampled every 1 ps from the MD trajectory . The autocorrelation function was computed as the dot-product of successive Φ–Ψ vectors using every 1 ps step as a new time origin and calculating the correlation function out to 200 ps . Several different time windows were used in the autocorrelation calculation , but all gave the same result . For comparison purposes , the same autocorrelation function was calculated for the first 118 Φ–Ψ angles ( out of 130 ) for an MD trajectory of a well-folded protein ( fibroblast growth factor 1; PDB ID = 1AXM ) . The MD trajectories of 45 F–F distances between the Cβ atoms were analyzed to calculate the probability distributions . Systematically , each of the F–F distance was interrogated and all of the distances were binned ( 1 Å bins from 0 to 60 Å ) to form a histogram of distance distributions . Probability distributions were calculated for each of the twenty simulations independently , and the values obtained were averaged at the end . A similar procedure was adopted for the mutant FG domain where the distance between the Cβ atoms of the Ala residue was used . Final probability distributions were used without any normalization . As a first approximation , the probability distributions were fit to a Gaussian distribution ( probability versus distance ) . This is a conservative approach and is expected to be valid considering the number of structures generated ( 60 , 000 ) during the molecular dynamics simulations and in the absence of any constraints . The center of the Gaussian is considered as the mean distance between the F–F ( or A–A ) , while the width at half-maximum is used as the allowed variation in the constraint . Clustering analysis: The correlation between different F–F probability distribution reflects the degree to which these variables ( F–F distances ) are related . The most common measure of correlation is the Pearson Product Moment Correlation ( http://www . r-project . org/ ) and reflects the degree of linear relationship between the two variables . In order to determine whether probability profiles of the F–F interaction correlate , a similarity matrix with a Pearson square metric was calculated . The correlation was used to indicate the presence ( or absence ) of relationship between various F–F interactions . The coding sequence for the representative 111 AA Nup116 FG domain was amplified from genomic S . cerevisiae DNA using PCR and was cloned into the vector pGEX-2TK in frame with the coding sequence for glutathione S-transferase ( GST ) at the 5′ end , and in frame with the coding sequence for six contiguous histidines at the 3′ end . Site directed mutagenesis was then used to alter the coding sequence for the mutant F>A FG domain . The correct coding sequences were confirmed by DNA sequence analysis . The FG domains were expressed in a E . coli BL21+ strain as fusion proteins with GST ( glutathione S transferase ) at the N-terminus and a HIS tag ( six contiguous histidine residues ) at the C-terminus . Glutathione coated Sepharose beads were then used to isolate each GST-FG domain fusion from crude bacterial cell extracts . The isolated FG domains were eluted from the beads by specific thrombin proteolysis of the GST tag . Nickel-coated beads were then used to capture and isolate the FG domain through its C-terminal His-tag , and the captured proteins were eluted from the beads using imidazole . Finally , the eluates were concentrated in a Centricon 3 unit and were size fractionated in an FPLC Superdex 200 sizing column that was equilibrated in 50 mM NaH2PO4 , pH of 6 . 5 for the NMR analysis , or in an FPLC Superdex 75 column equilibrated in 20 mM Hepes , pH 6 . 8 , 150 mM KOAc , 2 mM Mg ( OA ) 2 for determination of Stokes radii . Tandem-affinity purified wild-type and F>A mutant Nup116 FG domains were subjected to size-fractionation through an analytical-scale FPLC Superdex 75 column . FG domains ( 100 µl of 7 . 5 mg/ml ) were injected at a flow rate of 0 . 5 ml/min at 4°C into a column that was preequilibrated in 20 mM Hepes pH 6 . 8 , 150 mM KOAc , 2 mM Mg ( OAc ) 2 , and 0 . 5 ml fractions were collected . The FG domain elution profiles were monitored by UV absorbance at 280 nm and by SDS-PAGE analysis of the eluates . The nup elution profiles were compared to those of carbonic anhydrase ( 29 kDa , Rs = 23 . 5 Å ) , ovalbumin ( 45 kDa , Rs = 29 . 8 Å ) , and BSA ( 68 kDa , Rs = 35 . 6 Å ) , which served as molecular size standards . The elution volume of the standards was plotted in relation to their Stokes radii , allowing for estimation of the FG domain Stokes radii from the resulting linear regression formula . NMR experiments were performed on tandem-affinity purified FG domains dissolved in 50 mM NaH2PO4 , pH 6 . 5 . Final protein concentrations were ∼0 . 5 mM for both wild-type and mutant FG domains . NMR experiments were performed in a Varian INOVA 600 MHz spectrometer equipped with a 5 mm probe with a single-axis ( along Z ) shielded magnetic field gradients . One dimensional 1H NMR experiments were obtained using the water suppression scheme 1-3-3-1 Water-gate [61] . Self-diffusion coefficient measurements were obtained using a BPP-SED ( bipolar-gradient pulse pair selective echo dephasing ) sequence [62] . Translational diffusion tensor values were calculated based on the beads-model approximation of García de la Torre and Bloomfield [63] . This method has been used successfully to calculate translational as well as rotational diffusion tensors of proteins [40] , [64] . All atoms were considered as beads of equal size ( σ = 5 . 1 Å ) . The overall isotropic translational self-diffusion coefficient was calculated by taking the average of the principal values of the diffusion tensor . The hydrodynamic radius ( Rh ) for the wild-type and mutant Nup116 FG domains was calculated from the radius of gyration ( Rg ) values obtained from the simulations using the scaling relationship given in [39] . For native proteins , the scaling relationship is Rh = Rg/0 . 77 , and for proteins in strong denaturing conditions , the scaling relationship is Rh = Rg/1 . 06 . For the wild-type and mutant Nup116 FG domains simulated at 300 and 325 K , the hydrodynamics radius was obtained by Rh = Rg/0 . 77 . In the 350 K simulations , some of the protein conformations were highly extended ( as in denaturing conditions ) and a single scaling value was not appropriate . In this case , if the Rg for a structure was less than 30 . 7 Å for wild-type and 29 . 6 Å for the mutant , it was scaled by 1/0 . 77; if the Rg was greater , the value was scaled by 1/1 . 06 . The Rg cutoff values of 30 . 7 Å ( wild-type ) and 29 . 6 Å ( mutant ) were obtained by using Uversky's relationship: Rh ( 8 M urea ) = ( 0 . 22 ) *M0 . 52 , where M is the molecular mass [41] . The molecular mass for the simulated wild-type FG domain was 11 , 791 Daltons and for the mutant domain was 11 , 030 Daltons . The calculated Rh values were 22 . 5±4 . 0 for the wild-type FG domain and 28 . 6±5 . 2 for the mutant . To compare these Rh values to the Stokes radii values ( Rs , same as Rh ) for the purified FG domains in sieving columns , the contribution of a C-terminal His-tag ( 6 histidine residues/841 Da ) , which was added ( post simulations ) to the FG domains to aid in the purification of only full-length FG domains , had to be factored in . This was done using Uversky's scaling relationship by calculating Rs for the FG domains with the additional tag assuming a native pre-molten globular configuration for the wild-type and a native coil configuration for the mutant ( see Table 1 ) . The Rs estimated from the molecular dynamics simulations for the wild-type and mutant FG domains were multiplied by the ratio ( Rs ( His-tag ) / Rs ( no-tag ) ) to yield the final values of 23 . 1 ( ±4 . 1 ) Å for the wild-type FG domain and 29 . 6 ( ±5 . 4 ) Å for the mutant FG domain reported in Table 1 .
|
The nuclear pore complex is a molecular filter that gates macromolecular exchange between the cytoplasm and the nucleoplasm of cells . It contains a size-selective diffusion barrier at its center composed of proteins named FG nucleoporins . These nucleoporins feature large , structurally disordered domains that are highly decorated with phenylalanine–glycine ( FG ) sequence motifs . The dynamic structure of these disordered FG domains excludes them from classical structural biology analyses such as X-ray crystallography; thus , new approaches are needed to characterize their shape . Here computational and biophysical approaches were used to elucidate the ensemble of structures adopted by the FG domain of a nucleoporin . The analyses showed that the FG motifs function as intramolecular cohesion elements that compact the shape of the FG domain , forcing it to adopt loosely knit globular configurations that are constantly reconfiguring . Within the nuclear pore complex , dozens of these nucleoporin FG domains may stack as loosely knit globules forming a porous sieve that gates molecular diffusion by size exclusion .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"mathematics/statistics",
"computational",
"biology/molecular",
"dynamics",
"biophysics/macromolecular",
"assemblies",
"and",
"machines",
"cell",
"biology/nuclear",
"structure",
"and",
"function"
] |
2008
|
Intramolecular Cohesion of Coils Mediated by Phenylalanine–Glycine Motifs in the Natively Unfolded Domain of a Nucleoporin
|
The role of Hsp70 chaperones in yeast prion propagation is well established . Highly conserved Hsp90 chaperones participate in a number of cellular processes , such as client protein maturation , protein degradation , cellular signalling and apoptosis , but little is known about their role in propagation of infectious prion like aggregates . Here , we examine the influence of Hsp90 in the maintenance of yeast prion [URE3] which is a prion form of native protein Ure2 , and reveal a previously unknown role of Hsp90 as an important regulator of [URE3] stability . We show that the C-terminal MEEVD pentapeptide motif , but not the client maturation activity of Hsp90 , is essential for [URE3] prion stability . In testing deletions of various Hsp90 co-chaperones known to bind this motif , we find the immunophilin homolog Cpr7 is essential for [URE3] propagation . We show that Cpr7 interacts with Ure2 and enhances its fibrillation . The requirement of Cpr7 is specific for [URE3] as its deletion does not antagonize both strong and weak variant of another yeast prion [PSI+] , suggesting a distinct role of the Hsp90 co-chaperone with different yeast prions . Our data show that , similar to the Hsp70 family , the Hsp90 chaperones also influence yeast prion maintenance , and that immunophilins could regulate protein multimerization independently of their activity as peptidyl-prolyl isomerases .
Similar to human prion protein ( PrP ) , many yeast proteins have a tendency to undergo conformational conversion into infectious protein particles known as prions [1 , 2] [3] . The yeast prions have served as a great model system to understand the mechanistic details of prion induction , propagation , formation of prion variants and roles of various cellular factors in prion maintenance [4 , 5 , 6] . Two of the most extensively studied yeast prions [PSI+] and [URE3] are formed of native proteins Sup35 and Ure2 respectively . Sup35 is a translation termination factor and Ure2 is involved in nitrogen catabolism to repress nitrogen uptake from poor nitrogen sources , like proline , when a good nitrogen source such as ammonia , glutamine or asparagine , is present . Though the prion-forming domains of yeast proteins do not show a significant amino acid sequence similarity with many amyloid-forming proteins in mammals , various constituents of cellular machinery that influence yeast prions also affect mammalian amyloid based disorders , suggesting a functional conservation of cellular factors with regard to amyloidosis . Among various cellular factors , heat shock proteins ( Hsps ) , consisting mainly of the Hsp70 family and its co-chaperones , play a major role in the maintenance of yeast prions [7 , 8 , 9 , 10 , 11 , 12] . Hsp104 , which acts in coordination with the Hsp70 system , is essential for propagation of yeast prions . Also , Hsp104 overexpression leads to [PSI+] curing [13 , 14 , 15] . When present as the sole source of Ssa Hsp70 , Ssa1 antagonizes [URE3] , but not [PSI+] , while cells expressing only Ssa2 show a weak [PSI+] phenotype [16] . The overexpression of the Hsp70 co-chaperone Ydj1 or the Hsp70 nucleotide exchange factor ( NEF ) Sse1 cure [URE3] by a mechanism that requires interaction with Hsp70 [17 , 18] . Both Sse1 and Fes1 are required for [URE3] , and Sse1 also regulates [PSI+] induction and propagation in the prion variant specific manner [10 , 17 , 19] . Similarly , [SWI+] another well studied yeast prion is highly sensitive to alteration in activity of Hsp70 and its co-chaperones [20] . It is believed that chaperones , primarily Hsp70s , play crucial roles in promoting fibril growth and replication required for stable prion propagation [21] [22 , 23] [16] . Though the role of Hsp70 proteins in yeast prion maintenance has been widely studied , not much is known about Hsp90 function in this regard . The highly conserved Hsp90 family is essential for cellular growth in all eukaryotes . Hsp90 interacts with about 10% of all cellular proteins and the major clients include kinases , growth hormone receptors [24] , transcription factors [25] [26] and signal-transduction factors [27] [28] . Each protomer of dimeric Hsp90 consists of three domains: the N-terminal nucleotide binding domain ( NBD ) , middle domain ( MD ) , and a carboxy-terminal domain ( CTD ) . The Hsp90 dimer exists in a dynamic equilibrium between its open and closed conformation . In the open state Hsp90 is dimerized at its C-terminal domain with N-terminal domains separated . The large-scale conformational changes in Hsp90 upon ATP binding result in dimerization of N-terminal ATPase domains trapping substrate in the substrate binding pocket [29 , 30] . The charged linker region between the amino terminal and middle domains is crucial for dimerization of Hsp90 at the amino-terminal domain [31] . Hsp90 function is regulated by various other proteins that either modulate its ATPase activity or facilitate interaction with its client proteins . Hsp90 co-chaperones can be broadly divided into those with or without tetratricopeptide repeat ( TPR ) domains . The TPR domain containing co-chaperones such as Sti1 , Cpr6 , Cpr7 , Cns1 , Ppt1 and Tah1 compete for binding at the MEEVD motif present at the C-terminus of Hsp90 [32 , 33 , 34] and thus regulate different steps of Hsp90 reaction cycle . The deletion of the highly conserved MEEVD pentapeptide does not cause a growth defect in S . cerevisiae suggesting that even in the absence of the motif , Hsp90 is functional enough to support essential in vivo roles required for cellular viability . The TPR proteins participate in many cellular processes that include chaperonin activity , phosphatases , transcriptional regulation and cell cycle control [33] . The TPR domain containing Hsp90 co-chaperones influence client protein maturation by regulating Hsp90 ATPase activity , interaction with other cellular partner proteins , and conformational transition among various dynamic states formed during different stages of its reaction cycle . In addition to the TPR domain , Cpr6 and Cpr7 of the cylophilin family also contain a peptidyl-prolyl isomerase ( PPIase ) domain that catalyzes the cis-trans isomerization of peptide bonds N-terminal to the proline residues in proteins . Deleting the PPIase domain of Cpr7 does not affect yeast growth or Hsp90 activity , and the exact role of its PPIase domain in client maturation remains uncertain [35] . Unlike Cpr6 , Cns1 , which is essential for yeast cell growth , complements some Cpr7 functions indicating partial functional redundancy among some TPR proteins . Recently , there is emerging interest in the immunophilin family of proteins that includes FK506 binding proteins and cyclophilins for their role in various neurodegenerative diseases [36] [37] . Many Hsp90 clients first enter into the Hsp70-Hsp40 reaction cycle [38] . The bridge proteins Hop/Sti1 form an intermediate complex with Hsp70 and Hsp90 thus facilitating substrate transfer to Hsp90 , which remains in an ATP-free open conformation in the complex [39] . The Hsp90 cycle then begins with ATP binding and displacement of Hop/Sti1 with other TPR containing co-chaperones belonging to peptidyl-prolyl-cis/trans isomerase family such as yeast Cpr6 and Cpr7 [34] . Other non-TPR co-chaperones such as p23/Sba1 and Aha1 influence client maturation by modulating Hps90 conformational changes and its ATPase activity [40 , 41 , 42] . What determines substrate specificity and how a large pool of substrates gets partitioned between Hsp70 and Hsp90 is poorly understood . In contrast to the known role of cytosolic Hsp70s , not much is known about the role of Hsp90 and its co-chaperones on yeast prion formation and propagation . Though Hsp90 inhibition has no effect on [PSI+] , indirect evidence suggests some unknown role of Hsp90 proteins in yeast prions maintenance . Alterations of heat shock transcription factor , whose activity is regulated by Hsp90 , influence de novo [PSI+] formation and propagation [43] . Similarly , Hsp90 co-chaperones Sti1 and Cpr7 are important for [PSI+] curing upon Hsp104 overexpression [44 , 45] . Furthermore , Cpr7 is involved in maintaining conformation of prion variants [46] . Similarly , other TPR co-chaperone such as overexpressed Sgt2 , which is involved in the guided entry of tail-anchored proteins ( GET ) trafficking pathway , enhances excess Ssa Hsp70 mediated protection of [PSI+] from curing by excess Hsp104 [47] . Here , we explore whether Hsp90 plays any role , direct or indirect , on yeast prion propagation . We show that [URE3] propagation is not affected by inhibition of Hsp90 or by deleting either of the genes encoding Hsc82 or Hsp82 . Interestingly , though mutations affecting Hsp90 conformational changes have no effect on [URE3] , those lacking the C-terminal pentapeptide motif destabilize [URE3] , revealing that protein-protein complexes of Hsp90 and its TPR co-chaperones has a crucial role in prion maintenance . Our data suggest that the decrease in [URE3] stability in cells expressing Hsp90ΔMEEVD is due to loss of Hsp90 interaction with Cpr7 . Our study uncovers a role of Cpr7 in [URE3] propagation and that it is unique among known Hsp90 co-chaperones in its being required for prion maintenance . Our data suggest that Hsp90 controls the fate of not only its client proteins but also other cellular substrates through the functional modulation of other members of the chaperone machinery .
It is known that Hsp70 and Hsp90 family proteins influence folding and maturation of many cellular proteins , yet what determines chaperone specificity is not entirely clear . Whether yeast prions that are well known Hsp70 substrates also require Hsp90 remains unclear . In order to explore the role of Hsp90 chaperones on yeast prions , S . cerevisiae strains having deletion of either HSC82 or HSP82 were constructed . As seen in Fig 1A , no significant effect on [URE3] was observed suggesting that the presence of either Hsp90 is enough to support stable prion propagation . Alternatively , the Hsp90 family of proteins may be dispensable for [URE3] propagation . To address the latter possibility , we cultured yeast [URE3] cells lacking HSC82 ( strain SY297 ) in the presence of the known Hsp90 inhibitor 17-AAG ( S1 Fig ) . The inhibitor competes with ATP for binding to Hsp90 and thus inhibits its activity [48] . Inhibition of Hsp90 leads to Hsp70 overexpression and a defect in the maturation of its client proteins [49] . We first examined the effect of 17-AAG on Hsp70 expression and the steady state level of the Hsp90 client v-Src in yeast strain SY136 . The SY136 strain was transformed with plasmid ( pRS316PGAL-FLAG-vSrc ) encoding for FLAG-tagged v-Src . The transformants were further cultured in liquid SGal media lacking uracil in the presence and absence of 17-AAG . As expected , incubation with 17-AAG ( 50μM in DMSO ) upregulated Hsp70 expression . Also 17-AAG treated cells showed reduced steady state level of Hsp90 client protein v-Src ( S1A Fig ) with no significant change in its mRNA levels ( S1B Fig ) . However 17-AAG ( 100μM in DMSO ) mediated inhibition of Hsp90 had no effect on [URE3] stability , suggesting the Hsp90 function crucial for client maturation does not contribute much to [URE3] propagation ( S1C Fig ) . Hsp90 is essential and thus to examine whether Hsp90 influences [URE3] , we monitored prion propagation in yeast cells expressing the previously known Hsp90 mutants , His6-Hsp82Δ211–264 or His6-Hsp82ΔMEEVD as the sole source of Hsp90 [34] . Amino acids 211–264 form a part of the charged linker region required for Hsp90 conformational changes and thus cells expressing His6-Hsp82Δ211–264 as the only Hsp90 grow poorly at 37°C [34] . When combined with cpr7Δ the growth defects are additive even at 30°C . Also , strains expressing a similar construct Hsp90Δ211–263 show reduced activation and lower steady state level of its client v-Src [31] . The highly conserved C-terminal pentapeptide MEEVD is a common interaction site where various Hsp90 co-chaperones possessing TPR domains compete for binding to Hsp90 . As seen in Fig 1B , a similar frequency ( >99% ) of [URE3] colonies was observed in cells expressing wt His6-Hsp82 or His6-Hsp82Δ211–264 . However , cells expressing His6-Hsp82ΔMEEVD exhibit an increased [ure-o] phenotype as seen by an increase in the number of red colonies ( about 25 ± 5% ) , which arise from cells that lost [URE3] . The increase in frequency of [ure-o] colonies is not due to differences in steady state abundance of Hsp90 , Hsp70 , Ydj1 , or Sse1 , as shown in S2 Fig . The results indicate that the presence of the C-terminal MEEVD motif is required for the stability of [URE3] . Overall the data suggest an important previously unknown role of Hsp90 , independent of its client maturation activity , in the propagation of prions . As compromising Hsp90 client maturation activity does not affect [URE3] , the appearance of [ure-o] cells in the His6-Hsp82ΔMEEVD strain points toward a crucial role of Hsp90 co-chaperones in prion propagation . In order to examine the role of various Hsp90 co-factors , we created many single-knockout S . cerevisiae strains each lacking an Hsp90 co-chaperone ( Sti1 , Sba1 , Cpr6 , Cpr7 , Ppt1 , Tah1 , Hch1 or Aha1 ) and monitored [URE3] . As shown in Fig 2A the deletion of non-TPR Hsp90 co-chaperones ( Sba1 , Hch1 or Aha1 ) had no effect on the prion propagation . Among the TPR domain containing proteins , only deletion of the gene encoding Cpr7 profoundly affects [URE3] stability as seen by the appearance of red colony color as well as poor growth on solid medium lacking adenine ( Fig 2A and 2B ) . To examine whether this effect was [URE3] specific , we further constructed similar single-knockouts in strain 779-6A propagating the other well studied yeast prion [PSI+] . In contrast to its requirement for [URE3] , lack of Cpr7 had no effect on [PSI+] stability which is also in agreement with a previously reported study [50] . Similarly , other knockout strains carrying a single gene deletion encoding one of seven other Hsp90 co-chaperones also show no apparent differences in [PSI+] stability ( Fig 2C ) . We also examined the effect of deleting Cpr7 on strong and weak [PSI+] variants and found no difference in the prion stability in the presence and absence of Cpr7 ( S3 Fig ) . Thus both weak and strong [PSI+] variants remain largely unaffected by Cpr7 deletion which is distinct from the effect of Hsp70 co-chaperone Sse1 where deletion of Sse1 destabilize weak [PSI+] but has no effect on strong [PSI+] prions [51] . As described in Materials and Methods , the red phenotype in [ure-o] cells is due to repression of ADE2 transcription by Ure2 protein . Any cellular activity that negatively affects ADE2 transcription or Ade2 protein stability , even independently of Ure2 , could also produce similar red colony color phenotype as seen for [ure-o] . Thus , it was possible that [URE3] was still present in cpr7Δ cells , but its phenotype was masked by lack of Cpr7 . In order to explore these possibilities we transformed [ure-o] colonies from the cpr7Δ strain with a plasmid encoding Cpr7 ( pRS316PCPR7-CPR7 ) . If in cpr7Δ cells the red colony color is independent of [URE3] , or the [URE3] phenotype is masked , then the transformants would show white phenotype upon plasmid based complementation of CPR7 . Alternatively if the red phenotype of cpr7Δ cells is due to the absence of [URE3] , the transformants would continue to show red phenotype as the frequency of spontaneous appearance of [URE3] clones is rare . As shown in S4 Fig , more than 99% of transformants show red colony color phenotype suggesting deletion of gene encoding Cpr7 results in loss of [URE3] . In order to further examine whether [URE3] propagation depends upon the presence of Cpr7 , we sporulated a [URE3] diploid strain heterozygous for cpr7Δ ( CPR7/cpr7::KanMX ) and dissected multiple tetrads . In yeast the chromosomally inherited traits follow Mendelian segregation in 2:2 fashion following meiosis however the cytoplasmically inherited infectious yeast prion particles segregate in a non-Mendelian fashion and are transmitted to all four spores . Upon tetrad dissection of a diploid strain heterozygous for CPR7 , [URE3] segregated in a Mendelian 2:2 fashion , as seen by white colony color ( Fig 2D ) . These results suggest the [ure-o] phenotype is linked to one of the segregating genes . All colonies having red colony color , and none of the white colonies , grew normally on a plate containing G418 showing that all [ure-o] colonies lack gene encoding Cpr7 . Together our results clearly indicate that stable [URE3] maintenance requires the presence of Cpr7 and that the absence of [URE3] in cpr7Δ cells is not due to synthetic lethality of [URE3] with depletion of Cpr7 . Modulation of Hsp70 activity can alter yeast prion formation and propagation . Similarly , overexpressing the Hsp70 co-chaperones Ydj1 or nucleotide exchange factor Sse1 destabilize [URE3] . In order to explore whether the loss of [URE3] in cpr7Δ strain is due to alteration in the overall abundance of Hsp70 and/or its co-chaperones , their relative expression level was monitored in [URE3] and [ure-o] cells of wild type and cpr7Δ strains . As seen in Fig 3A , Ydj1 , Sse1 and Hsp90 are expressed at similar levels in both wild type and cpr7Δ strains , regardless of [URE3] status . In contrast , the Hsp70 level was increased by about 1 . 5–2 . 0 fold in the cpr7Δ strain as compared to that of wild type strains ( Fig 3B ) . The four Ssa Hsp70 isoforms in S . cerevisiae influence [URE3] propagation , and though they are highly homologous they have distinct effects on yeast prions [16] . In order to examine whether the decrease in [URE3] stability in cpr7Δ strain is due to an increase in Hsp70 , wild type strain SY187 was transformed with a plasmid ( pRS315PSSA2-SSA1/2/3or4 ) encoding one of the four Ssa Hsp70 isoforms . Nevertheless , the Leu+ primary transformants expressing an additional Ssa Hsp70 isoform show stable [URE3] phenotype similar to that seen for the wild type strain transformed with pRS315 ( Fig 3C ) . These results suggest that increase in Hsp70 upon CPR7 deletion does not contribute significantly to [URE3] destabilization . Additionally , we previously showed that further subculturing of transformants overexpressing Ssa1 destabilize [URE3] with about 7–8% of cells showing [ure-o] phenotype [52] . However , as the absence of Cpr7 leads to more than 99% of cells showing [ure-o] phenotype , the near complete loss of [URE3] in cpr7Δ cells can be interpreted as being caused by something other than an increase in amount of Ssa1 [52] . As a way to confirm that the anti-[URE3] effects of depleting Cpr7 were not due to increased expression of Ssa1 , we constructed strain NY17 by deleting the gene encoding Cpr7 in strain NY16 that expresses Ssa2 under Ssa2 promoter as the only source of Ssa Hsp70 . As seen by red colony color phenotype and poor growth on Ade- solid growth medium in S5A Fig , depleting Cpr7 caused loss of [URE3] . This loss occurred despite the complete absence of Ssa1 , Ssa3 and Ssa4 . We examined the abundance of Hsp70 in strains NY16 and NY17 and found it was relatively higher in the cpr7Δ strain ( S5B Fig ) . As elevated Ssa2 abundance supports a stable [URE3] phenotype ( S5C Fig ) , the increase in [ure-o] colonies in strain NY17 must be unrelated to an increase in amount of Ssa2 . The Hsp90 associated immunophilins Cpr6 and Cpr7 belong to the PPIase family of enzymes and share more than 50% sequence similarity , yet they can act in distinct cellular processes [53] [54] . In order to examine whether Cpr6 can complement the role of Cpr7 with regard to [URE3] , the cpr7Δ strain harbouring Cpr7 on a URA3-based plasmid ( pRS316PCPR7-CPR7 ) was transformed with a plasmid encoding Cpr6 or Cpr7 ( pRS413PTEF-CPR6 or pRS413PTEF-CPR7 , respectively ) regulated by the TEF promoter . We then used FOA selection to isolate cells having lost the resident URA3 plasmid . The resulting cpr7Δ strains with plasmids encoding Cpr6 or Cpr7 were grown in liquid medium selecting for plasmid maintenance and then spread onto ½ YPD plates . Colony color was noted after incubation at 30°C for 3–4 days , at which time the ½ YPD plates were replicated onto SD medium lacking histidine ( for monitoring plasmid loss ) or adenine ( for monitoring [URE3] ) . As shown in Fig 4B , cells expressing Cpr7 colonies were white on ½ YPD and grew well on medium lacking adenine , suggesting that the plasmid expressed Cpr7 functionally complements the chromosomal knockout of CPR7 with regard to [URE3] . In contrast , cells harbouring either an empty plasmid ( ev ) or a plasmid encoding Cpr6 show similar red colony color on ½ YPD . After continued incubation on adenine deficient growth medium at 30°C , cells expressing Cpr6 grew only poorly , which is similar to cells with the empty plasmid . Thus , Cpr6 was unable to complement Cpr7 function with regard to [URE3] maintenance . The amino acid identity between the PPIase and TPR domains of Cpr6 and Cpr7 is about 45% and 32% respectively . To examine which domain of Cpr7 is crucial for [URE3] propagation , hybrid proteins , based upon the Cpr6 and Cpr7 as the parent proteins , were constructed in which either the PPIase ( 6PPI/7TPR ) or TPR ( 7PPI/6TPR ) domain of Cpr7 was swapped with that of Cpr6 ( Fig 4A ) [34] and their ability to support [URE3] was again monitored as described above for Cpr6 . As shown in Fig 4B , cells expressing 7PPI/6TPR show red colony color phenotype on ½ YPD medium . The 7PPI/6TPR expressing colonies show either no growth ( day 1 ) or poor growth ( day 2 ) on adenine deficient medium . In contrast , the frequency of white colonies on ½ YPD plate and growth on plates lacking adenine were found to be similar in strains expressing 6PPI/7TPR and wild type Cpr7 suggesting that the TPR domain of Cpr7 is crucial for [URE3] propagation . Furthermore , cells expressing a truncated derivative of Cpr7 lacking the PPIase domain also showed a stable [URE3] phentoype , suggesting peptide-prolyl activity of Cpr7 does not contribute much to the prion stability . Cns1 is unique among the known TPR containing Hsp90 co-chaperones in being essential for yeast cell growth . The 385 amino acid protein contains three TPR motifs and is a suppressor of the growth defect caused by cpr7Δ , which suggests partial functional redundancy between Cpr7 and Cns1 [55] . In order to explore whether Cns1 complements for the loss of Cpr7 with regard to [URE3] , a plasmid encoding Cns1 ( pRS426PCNS1-CNS1 ) was transformed into cpr7Δ cells pooled from Ade- medium and transformants were further monitored for [URE3] propagation as described above for Cpr6 . As seen before , cells expressing Cpr7 from the constitutive TEF promoter showed stable [URE3] on ½YPD ( Fig 5 ) . As expected , cells that lost the Cpr7-encoding plasmid ( no growth on histidine deficient medium ) show red colony color and no growth on Ade- medium . As we saw for Cpr7 , we found that cells that express Cns1 are white and grow normally on plates lacking adenine , suggesting that Cns1 functionally complements Cpr7 with regard to [URE3] propagation . Cpr7 belongs to the family of immunophilins that are also known to interact with many unfolded substrates to prevent their aggregation and keep them in a folding competent state [54] [56] . We used a pull-down assay using purified His6-Cpr7 as bait to examine interaction of Cpr7 with Ure2 . The cell lysate from wild type [ure-o] cells was fractionated and the supernatant was incubated with Cpr7-bound cobalt based metal affinity resin . The unbound proteins were washed and the bound fraction was probed by immunobloting with antibodies specific against Ure2 or Hsp90 . In agreement with previous studies , Hsp90 was identified in fractions bound to His6-Cpr7 ( Fig 6 ) . The deletion of MEEVD leads to a significant decrease in Hsp82 interaction with Cpr7 . Interestingly , Ure2 was also detected in the fraction that bound to His6-Cpr7 . The presence of Ure2 in this fraction suggests that Ure2 either interacts with Cpr7 or Hsp82 . We thus examined Ure2 interaction with Cpr7 using cell lysate from cells expressing Hsp82ΔMEEVD . If the presence of Ure2 in the fraction bound to Cpr7 is indirectly due to its interaction with Hsp82 , the loss of Cpr7-Hsp82 interaction upon deletion of C-terminus MEEVD motif of Hsp82 would lead to a decrease in Ure2 level in the bound fraction . As shown , however , a similar level of Ure2 was detected in the bound fraction obtained using lysates from cells expressing either Hsp82 or Hsp82ΔMEEVD , suggesting that Cpr7 interacts directly with Ure2 . To substantiate this conclusion , we also examined Cpr7 interaction with Ure2 in vivo by using cells expressing His6-Cpr7 . We co-transformed strain SY187 with plasmids expressing His6-Cpr7 ( pRS413PTEF-His6-CPR7 ) and Ure2-GFP ( pRS426PGPD-URE2-GFP ) or GFP ( pRS426PGPD-GFP ) . Transformants were grown further and the lysate was passed over cobalt based metal affinity resin . As shown in S6 Fig , Ure2 was detected in the bound fraction obtained only from the strain co-expressing His6-Cpr7 , again suggesting Cpr7 interacts with Ure2 in vivo . Overall , these results indicate that the role of Cpr7 in [URE3] maintenance could be mediated by its ability to interact directly with Ure2 . As Cpr7 interacts with Ure2 and is required for stable [URE3] propagation , we speculated that Cpr7 might have the ability to modulate Ure2 fibrillation required for stable [URE3] maintenance . In order to examine the effect of Cpr7 on amyloid forming tendency of Ure2 , we monitored fibrillation of Ure2 in vitro using Thioflavin T ( ThT ) , a dye that forms fluorescent complexes with amyloids , in the presence and absence of Cpr7 . As expected ThT fluorescence intensity increases upon incubation with Ure2 at 37°C , suggesting spontaneous fibrillation of Ure2 with a lag of about 2–3 min . Cpr7 and Cpr6 alone did not affect ThT fluorescence intensity . Pre-incubation of Ure2 with Cpr7 did not affect the lag time . However , the rate of amyloid formation and amyloid yield increased significantly , indicating that Cpr7 promotes Ure2 fibrillation . The increase in Ure2 fibrillation was observed only upon incubation with Cpr7 and not its near homolog Cpr6 ( Fig 7 ) . The increase in fluorescence intensity was not due to a direct effect of Cpr7 on ThT as fluorescence intensity was similar when similar amounts of pre-aggregated Ure2 protein , with and without Cpr7 , was measured ( S7A and S7B Fig ) . As the isolated TPR domain of Cpr7 supports [URE3] , we also tested if this domain influenced ure2 amyloid formation . Interestingly , as seen for Cpr7 , the addition of 7TPR to Ure2 similarly enhanced its fibrillation . These results are in agreement with the in vivo results that showed that 7TPR , but not Cpr6 , complemented Cpr7 function with regard to [URE3] propagation . Together these results support the conclusion that Cpr7 plays an important role in [URE3] propagation by enhancing Ure2 fibrillation .
Studies on yeast prions have formed the basis of understanding mechanistic aspects of the formation and propagation of not only of prions , but also other amyloid based disorders . The insight obtained has broadened our understanding not only of the role of various cellular factors in prion maintenance , but also of basic cellular biology . Though both Hsp70 and Hsp90 families of proteins are central to protein quality control in the cell , most of the cellular factors that influence prion propagation belong to the Hsp70 family and its co-chaperones . Whether prion propagation also requires Hsp90 machinery is not yet clear and the role of Hsp90 and its co-chaperones has been shown only to maintain prion conformation and the curing of [PSI+] upon overexpression of Hsp104 [44] [45] [46 , 50] . In the present study we show that Hsp90 , by virtue of its ability to interact with TPR proteins , is critical for yeast prion propagation . Furthermore we show that Cpr7 is a novel cellular factor required for [URE3] maintenance . Here we investigated the role of Hsp90 and its co-chaperones on [URE3] propagation . The knockout of genes encoding either Hsc82 or Hsp82 has no adverse effect on [URE3] . This is in contrast to the role of Ssa Hsp70 , as different Ssa Hsp70 isoforms show distinct effects on [URE3] [16] . Also , the inhibition of Hsp90 ATPase activity with 17-AAG did not alter [URE3] propagation . Together these results suggest that Hsp90 function specific for the maturation of its client proteins is not crucial for [URE3] maintenance . This is also in agreement with previous observations suggesting that Hsp90 is a more specialized chaperone than Hsp70 , which acts on a much broader range of substrates in the cell . The Hsp90 mutant lacking charged linker region ( amino acids from 211 to about 263 ) is defective in ATP dependent amino terminal dimerization , shows a lower steady state level of its client v-Src , and subsequent maturation of its client proteins such as glucocorticoid receptor and v-Src [31] . The strain expressing Hsp90Δ211–264 as sole Hsp90 source also supports a stable [URE3] propagation which provides further evidence that [URE3] stability does not require optimal Hsp90 activity . Hsp90ΔMEEVD , however , which lacks the binding motif for TPR containing co-chaperones , is incapable of propagating [URE3] stably when it is expressed as the only source of Hsp90 . It is possible that Hsp90 interaction with its TPR co-chaperones might modulate the latters' functions in a way that is required for maintenance of [URE3] . Alternatively , the loss of Hsp90 interaction might increase affinity of TPR co-chaperones of Hsp90 toward other cellular proteins having a binding motif similar to MEEVD and such new interactions might lead to [URE3] stability . Collectively the data suggest that although the Hsp90 chaperone cycle including client binding and release may not be involved in [URE3] maintenance , the chaperone might influence [URE3] propagation indirectly through the interaction of its C-terminus with partner co-chaperones that can influence [URE3] . The results from multiple single knockout strains of Hsp90 co-chaperones reveal that Cpr7 is unique among various Hsp90 co-chaperones with regard to its role in the maintenance of mitotically stable [URE3] variant in our yeast strain . Similar to many of the other yeast prions , variants of [URE3] also occur in S . cerevisiae , and whether or not the essential role of Cpr7 in [URE3] stability is extendable to the other [URE3] variants remains to be explored . Among the various Hsp90 co-chaperones that facilitate different steps of the Hsp90 reaction cycle , only Cpr7 deletion affects [URE3] , suggesting a specific and not a general function of Hsp90 could be essential for [URE3] stability . This is also supported by the observation that cells expressing Hsp82ΔMEEVD and not Hsp82Δ211–264 show [URE3] loss , which points toward a specific role of the C-terminus of Hsp90 in [URE3] propagation . Cpr7 deletion also alters nucleotide dependent Cpr6 interaction with Hsp90 , however , as Hsp82Δ211–264 , which also leads to a similar effect , does not destabilize [URE3] , the alteration of Hsp90-Cpr6 interaction could not be the cause of [URE3] loss from cpr7Δ cells [34] . Collectively the results suggest that Cpr7 influences [URE3] propagation directly by participating in processes required for efficient [URE3] propagation , rather than indirectly by influencing Hsp90 function . The possibility that Cpr7 impacts [URE3] directly is in agreement with previous reports showing that cyclophilins could interact with substrate directly to assist their folding and prevent aggregation . Similar to the role of Cpr7 in [URE3] propagation , another related immunophilin FKBP52 , present in brain , induces aggregation of the pathological mutant of Tau , Tau-P301L [36 , 57] , suggesting that the role of Cpr7 in [URE3] is not an isolated example of the significance of immunophilins in amyloidosis , but is widespread in higher eukaryotes including mammals . Yeast cells express several TPR domain containing proteins , but how these interact with their specific partner proteins and influence a specific function of the interacting partner is not clearly understood . Though Cpr7 is primarily an Hsp90 co-chaperone , previous studies suggests that Cpr7 also interacts with Hsp70 [50 , 58] . Similarly Cpr6 is also known to interact with Hsp70 proteins . Similar to other Hsp90 TPR co-chaperones , such interaction could be mediated by the TPR domain which is known to interact with EEVD-like motifs . However in contrast to the known influence of several other TPR co-chaperones on Hsp70 activity or function , the significance of Cpr6 or Cpr7 interaction on wild type Ssa Hsp70 activity has still not been clearly shown [59] . Indeed , while Cns1 stimulates Ssa1 ATPase activity by about 8-fold , no stimulation was seen by Cpr7 [60] . Similarly , another Hsp90 co-chaperone Sti1 that interacts with Hsp70 strongly stimulates Ssa Hsp70 ATPase activity and helps in the substrate transfer from Hsp70 to Hsp90 [61] . Since our study shows that Cns1 complements Cpr7 function with regard to [URE3] , and Cns1 , but not Cpr7 , stimulates Hsp70 activity , it is more likely that such a profound effect of cpr7Δ on [URE3] stability is not mediated entirely through Hsp70 . This idea is further supported by the fact the Cpr7 influences Ure2 fibrillation in the absence of any additional cellular factor . Among the various knockout strains , although cpr7Δ decreased [URE3] stability significantly , the lack of effect by its nearly homologous cyclophilin Cpr6 suggests a clear distinction between in vivo roles of these closely related Hsp90 associated cyclophilins . This is also in agreement with previous reports showing non-redundancy in Cpr6 and Cpr7 functions such as their ability to support yeast growth [53] . By using Cpr6 and Cpr7-based hybrid proteins , we find that the TPR domain acts as the regulatory domain to govern functional distinctions between Cpr7 and Cpr6 with regard to [URE3] . As the PPIase domain of Cpr7 is dispensable for [URE3] , and a hybrid protein containing the TPR domain from Cpr7 supports stable [URE3] propagation , the distinct action of the two Hsp90 co-chaperones could be due to their different chaperone activity independent of the PPIase function . Stable prion propagation requires continuous growth and breakage of fibrils to generate more seeds for further fibril growth and transmittance to daughter cells . Previous studies show that Cpr7 has more potent chaperone activity than Cpr6 , thus it is possible that Cpr7 might be more efficient to help break larger fibrils into smaller ones and hence able to generate more seeds required for [URE3] propagation . Another TPR co-chaperone , Cns1 restores some Cpr7-dependent activities such as the effect on cell growth , negative regulation of heat shock factors ( HSF ) and maturation of glucocorticoid receptor [55 , 62] . Our study now reveals that overexpressed Cns1 also complements Cpr7 function for [URE3] propagation . The HSF and glucocorticoid receptor require Hsp90 and thus Cpr7 deletion could affect these substrates through modulation of the Hsp90 reaction cycle . It is known that Cns1 restores altered Cpr6-Hsp90 interaction upon loss of Cpr7 [34] , however , as discussed above , the complementation of Cpr7 function by Cns1 for [URE3] is not due to restoration of Cpr6 interaction with Hsp90 , but to redundancy in other Cpr7 functions required for [URE3] stability . As indicated above , Hsp90 client binding function is not crucial for [URE3] stability and the effect of Cpr7 deletion on [URE3] was not caused by altering Hsp90 interaction with Cpr6 . Also , we did not find a direct interaction of Hsp90 with Ure2 , but did see a direct influence of Cpr7 on Ure2 fibrillation in the absence of Hsp90 . Collectively , our results suggest that it is most likely that the effect of Cpr7 deletion on [URE3] is independent of Hsp90 and due to a loss of specific function of Cpr7 in cpr7Δ strains . Together our data further suggest that Cns1 could also perform functions independent of Hsp90 , and that functional redundancy between Cns1 and Cpr7 is not only limited to their roles as Hsp90 co-chaperones . In vitro , Cns1 is unable to prevent aggregation of citrate synthase , and also fails to promote refolding of unfolded RNase T1 , suggesting that unlike Cpr7 , Cns1 not only lacks PPIase activity but also chaperone function for some Cpr7 substrates [60] [54] . Though Cns1 fails to chaperone RNase T1 or Citrate Synthase , it might still be able to influence Ure2 aggregation . As Cns1 lacks PPIase activity , the restoration of [URE3] propagation upon Cns1 overexpression in cpr7Δ strain could be through its TPR domain . Though the TPR domain of Cpr7 is only about 10% identical with that of Cns1 , the functional complementation of Cpr7 by Cns1 shown in previous and present studies suggests that despite low identity , the structural region critical for Cpr7 function remains partially conserved between the two TPR proteins . The pull down assay using Cpr7 as a bait protein reveals a novel interaction of Cpr7 with Ure2 . Our in vitro ThT assay shows that in the absence of other additional cellular factors , Cpr7 is able to promote Ure2 fibrillation . Though the mechanism by which Cpr7 promotes Ure2 fibrillation is not clear , it is possible that similar to Hsp70s , Cpr7 may act as a chaperone to promote solubilisation of preformed amyloid and during the process breaks a larger fibril into smaller fibrils and thus generates more seeds for further fibrillation . Cpr7 thus might be essential for the prion replication required for stable transmittance of [URE3] prion seeds into daughter cells . Alternatively , in the absence of any additional factor Ure2 assembly could be partitioned between on-pathway and off-pathway intermediates , and the presence of Cpr7 might shift the process more towards the formation of on-pathway intermediates for amyloid formation . Also , as cells expressing the Hsp90ΔMEEVD mutant show [URE3] loss , it is possible that Hsp90 interaction with Cpr7 might modulate its chaperone activity either through inducing conformational changes in Cpr7 or yet other unknown mechanism required for [URE3] maintenance . Hsp90 is previously known to induce conformational changes of its co-chaperone Ppp5 that further activates its phosphatase activity [63] . Alternatively , the retargeting of Hsp90 TPR co-chaperones in the Hsp90ΔMEEVD strain to other cellular factors possessing motifs similar to MEEVD might influence [URE3] stability . Thus , intact Hsp90 could be supporting [URE3] stability by sequestering Cpr7 from binding to its alternative partner protein that in turn might destabilize the prion . Our data reveal a novel role of cyclophilin Cpr7 in yeast prion [URE3] propagation . Thus , in addition to Hsp70 and its co-factors , Hsp90 and some of its co-chaperones also modulate the maintenance of this yeast prion . The role of immunophilins in amyloid based human diseases such as Alzheimer’s diseases , is emerging , and thus our finding that yeast immunophilins also modulate a yeast prion in a manner required for its efficient propagation provides a genetically tractable system for further understanding the role of immunophilins in amyloidosis . Why [PSI+] does not require TPR co-chaperone Cpr7 remains to be investigated . Interestingly , cellular factors required for stable [URE3] maintenance , such as Ssa2 , are not essential for [PSI+] . Similarly , Hsp104 overexpression efficiently cures [PSI+] but not [URE3] . Though more is needed to be explored , it is tempting to speculate that the distinct chaperone requirement for [URE3] and [PSI+] could be due a specific requirement of Hsp90 chaperone machinery in [URE3] stability .
Strains and plasmids used are described in Tables 1 and 2 , respectively . The wild type strain SY187 ( MATa , kar1-1 , PDAL5::ADE2 , his3Δ202 , leu2Δ1 , trp1Δ63 , ura3-52 ) encodes ADE2 gene regulated by the DAL5 promoter for monitoring [URE3] . The cpr6Δ ( NY14 ) and cpr7Δ strains ( NY01 ) were constructed by genomic integration of cpr6::KanMX or cpr7::KanMX cassette , respectively , into SY187 using homologous recombination . The hsp82Δ ( SY295 ) and hsc82Δ ( SY297 ) strains were constructed using strain MR195 or MR194 ( Kind gift from Dr . Masison’s lab ) , respectively , using standard yeast genetic manipulations . Strain NY02 containing double deletion of HSC82 and HSP82 and harbouring plasmid pRS316PHSP82-HSP82 was constructed from SY295 using standard yeast genetic methods . The strains NY03 , NY04 , NY05 , NY06 and NY07 were made by shuffling pRS316PHSP82-HSP82 in strain NY02 with pRS413PTEF-His6-HSP82 , pRS413PTEF-His6-HSP82 ( ΔMEEVD ) , pRS413PTEF-His6-HSP82 ( Δ211–264 ) , pRS413PTEF-HSP82 and pRS413PTEF-HSP82 ( ΔMEEVD ) , respectively . The knockout strain for other Hsp90 co-chaperones ( Sti1 , Sba1 , Cpr6 , Cpr7 , Ppt1 , Tah1 , Hch1 or Aha1 ) was constructed as described above for the cpr7Δ strain using standard genetic techniques . The genes encoding Cpr6 or Cpr7 were cloned under their respective native promoters in pRS316 to construct pRS316PCPR6-CPR6 or pRS316PCPR7-CPR7 , respectively . Similarly , pRS413PTEF-CPR6 and pRS413PTEF-CPR7 were constructed to express Cpr6 or Cpr7 respectively from the TEF promoter . Overlap PCR was carried out to construct genes encoding hybrid proteins 6PPI/7TPR or 7PPI/6TPR . The 6PPI/7TPR encodes amino acids ( a . a . ) 1–176 of Cpr6 and 198–393 of Cpr7 . Similarly 7PPI/6TPR encodes a . a . 1–197 of Cpr7 and 177–371 of Cpr6 . The full length PCR amplified product was further digested with BamHI and XhoI , and subcloned into plasmid pRS413PTEF . Similarly plasmid pRS413PTEF-7TPR , encoding the TPR domain of Cpr7 ( a . a . 193–393 ) ( henceforth 7TPR ) , was constructed by restriction digestion ( BamHI and XhoI ) followed by ligation of a PCR amplified gene product encoding the TPR domain of Cpr7 into plasmid pRS413PTEF digested with similar restriction enzymes . For immunoblot and protein purification of Cpr7 and Cpr6 , a short tract of Hexa-His-tag was added at the 5’ end of gene encoding Cpr7 or Cpr6 using standard PCR-based recombinant technology . For expression in E . coli , pET29bHTV-CPR6 , pET29bHTV-CPR7 or pET29bHTV-7TPR was constructed to encode from 5’ to 3’ direction , a Hexa-His tag , and a TEV protease recognition site followed by the gene encoding CPR6 , CPR7 or 7TPR , respectively . Media and growth conditions are as described previously [64] . ½ YPD medium contains 0 . 5% yeast extract , 2% peptone , 2% dextrose and 2% agar . Liquid YPAD is similar to ½ YPD except it contains 1% yeast extract and 200mg/L adenine . SD and SGal are synthetic dextrose and synthetic galactose minimal media , respectively . Cells were grown at 30°C unless otherwise stated . The heterologous diploid ( CPR7/cpr7::KanMX ) was constructed by mating SY187 [URE3] MATα with NY01 MATa strain ( cpr7Δ[ure-o] ) and selecting on adenine deficient medium containing G418 ( 200 μg/ml ) . Diploids were further streaked to pure colonies on ½ YPD plates and further confirmed by replica plating onto adenine deficient SD plate containing G418 . Diploids were sporulated on a 2% potassium acetate plate and tetrad dissection was carried out on ½ YPD . Gln3 is an activator of DAL5 promoter . When cells are grown in the presence of a good nitrogen source such as standard ammonium containing media , Ure2 binds to transcription factor Gln3 and represses the DAL5 promoter . To monitor [URE3] , our strains encode ADE2 controlled by the DAL5 promoter . In [ure-o] ( lacking [URE3] ) cells , functional Ure2 represses transcription of ADE2 so cells do not grow in the absence of adenine and remain red on limiting adenine media . [URE3] cells contain predominantly inactive Ure2 , which relieves repression of DAL5 and allows expression of Ade2 . Thus [URE3] cells grow on media lacking adenine and remain white on limiting adenine media . Sup35 is a translation termination factor and thus its conversion to the functionally inactive [PSI+] form leads to suppression of ade2-1 nonsense allele . This suppression also requires the weak tRNA suppressor Sup16 ( encoded by SUQ5 ) , which alone does not suppress ade2-1 . Thus , in cells propagating [PSI+] , enhanced nonsense suppression of ade2-1 leads to the synthesis of Ade2 . In contrast , [psi-] cells lack functional Ade2 and thus are unable to grow in the absence of externally added adenine and remain red when grown on limiting adenine . The prion variants of [URE3] and strong [PSI+] in our strains are of unknown origin . Our strains 1566 and 1567 are kind gift from Dr . Daniel Masision , and contain the [PSI+]Sc4 and [PSI+]Sc37 variants respectively that were introduced by cytoduction from J . Weissman strains JW127 and JW129 [51] . For Cpr7 purification , the plasmid pET29bHTV-CPR7 was transformed into Escherichia coli strain Rosetta 2 ( DE3 ) ( Invitrogen ) . The pre-grown culture of optical density at 600nm ( O . D . 600nm ) ~0 . 6 was induced with isopropyl-β-D-thiogalactopyranoside ( IPTG ) at 18°C . Cells were lysed and the protein was purified from supernatant using cobalt based Talon metal affinity resin . Cpr7 was eluted with 300mM imidazole and further incubated overnight with His6-TEV ( molar ratio , Cpr7/His-TEV:20/1 ) protease at 4°C . The sample was extensively dialyzed and further incubated with cobalt metal affinity column and the His6-tag cleaved Cpr7 protein was collected as unbound fraction . The protein was further purified using anion exchange ( Mini Q ) chromatography . Protein purity was verified using SDS-PAGE and identity was confirmed using Mass spectrometry . Cpr6 and 7TPR were purified using the procedure described above for Cpr7 . The plasmid pKT41-Ure2 ( kind gift from Dr . Reed Wickner , National Institute of Health , Bethesda ) encodes Ure2 with an N-terminal Hexa-His-tag . The plasmid was transformed into E . coli strain Rosetta 2 ( DE3 ) and Ure2p was purified using cobalt based Talon metal affinity resin as described before [52] . Thioflavin T ( ThT ) ( 500μM ) was added to 48 μM of purified Ure2 with and without Cpr6 , Cpr7 or 7TPR ( 30 μM each ) in a 96 well plate . The plate was incubated at 37°C with a shaking speed of 900 rpm in linear mode in a multimode plate reader ( TECAN infinite M200 PRO ) . Fluorescence kinetics was measured after every 15 minutes with emission wavelength of 485 nm upon excitation at 450 nm . Each experiment was repeated at least three times . Cells were lysed using lysis buffer ( phosphate-buffered saline with 0 . 2% TritonX-100 and protease inhibitor cocktail ) by vortexing with glass beads and lysate was fractionated into supernatant and pellet . About 10 μg of total protein was separated on sodium dodecyl sulfate polyacrylamide gels and transferred onto polyvinylidene fluoride membranes . Anti His6-tag antibodies ( Pierce Biotechnology ) were used to capture His6-Cpr7 . For Pull-down assay , purified His6-Cpr7 was bound over cobalt based metal affinity resin and yeast lysate was passed through the resin . The resin was washed and the bound fractions were probed with desired antibodies .
|
Studies on yeast prions have provided insight into the role of various cellular factors involved in amyloid based disorders . Among the various cellular components the chaperone network has emerged as a crucial regulator of formation and propagation of yeast prions . The chaperone machinery primarily consists of Hsp70 and Hsp90 families of proteins , whose activity is further modulated by multiple co-chaperones . Though it is known that Hsp70s and its co-factors influence prion strength and stability , the role of Hsp90 chaperone machinery in yeast prion propagation is not clear . Here we examine the role of Hsp90 chaperones and show that the C-terminal MEEVD motif on Hsp90 , which is required for interaction with tetratricopeptide repeat domain containing co-chaperones , is crucial for propagation of the yeast prion [URE3] . Further study revealed a novel role of the Hsp90 co-chaperone Cpr7 in the propagation of infectious amyloid and thus broadens our understanding about the cellular role of immunophilins .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Hsp90-Associated Immunophilin Homolog Cpr7 Is Required for the Mitotic Stability of [URE3] Prion in Saccharomyces cerevisiae
|
Upon HIV transmission , some patients develop AIDS in only a few months , while others remain disease free for 20 or more years . This variation in the rate of disease progression is poorly understood and has been attributed to host genetics , host immune responses , co-infection , viral genetics , and adaptation . Here , we develop a new “relaxed-clock” phylogenetic method to estimate absolute rates of synonymous and nonsynonymous substitution through time . We identify an unexpected association between the synonymous substitution rate of HIV and disease progression parameters . Since immune activation is the major determinant of HIV disease progression , we propose that this process can also determine viral generation times , by creating favourable conditions for HIV replication . These conclusions may apply more generally to HIV evolution , since we also observed an overall low synonymous substitution rate for HIV-2 , which is known to be less pathogenic than HIV-1 and capable of tempering the detrimental effects of immune activation . Humoral immune responses , on the other hand , are the major determinant of nonsynonymous rate changes through time in the envelope gene , and our relaxed-clock estimates support a decrease in selective pressure as a consequence of immune system collapse .
Although the clinical course of HIV infection is generally well-defined , there is considerable variability among patients in rates of disease progression . The highly variable asymptomatic phase , ranging from several months to more than 20 years , most likely reflects differences in the nature of the evolutionary arms race between the virus population and the host immune system . Both humoral and cell-mediated immune responses are mounted against the virus but are eventually defeated by HIV replication and adaptation . As part of this process , neutralizing antibodies ( nAbs ) exert a strong selective pressure on the HIV envelope gene ( env ) [1 , 2] but do not control viral replication , and nAb levels are not predictive of disease progression [3] . Cytotoxic T cell ( CTL ) responses play a more important protective role in HIV infection , and evidence has shown that partial control of HIV replication in vivo is temporally associated with the appearance of CTL responses [4] and that the rate of disease progression is strongly dependent on human leukocyte antigen ( HLA ) class I alleles [5 , 6] . Although CTLs may not be responsible for the majority of infected cell deaths , small differences in CTL killing rates could still be clinically relevant and alter the time of disease onset [7] . Unfortunately , the enormous potential for evolutionary change in HIV can counteract these host defense responses . High mutation and recombination rates coupled with rapid replication dynamics generate a genetically diverse viral population , enabling the infection of a large number of susceptible cells . Consequently , HIV is able to adapt readily to changing environmental conditions within each host . The envelope protein is able to evade nAb responses by accumulating multiple amino acid changes , especially in the hypervariable regions , while maintaining full functionality for viral cell entry . CTL responses , however , target epitopes in other viral genes ( such as gag and nef ) more strongly and/or more frequently , particularly during the initial stages of HIV infection [8–10] . Although HIV infection often results in the evolution of viral variants that escape CTL responses [11] , recent evidence from transmission pair studies and in vitro growth rate studies suggests that some escape mutations might occur at the expense of viral fitness [12 , 13] . Although the immune response against HIV has become increasingly well-characterized , less is known about the role of viral evolution in disease progression , despite its importance to our understanding of virus–host dynamics during persistent infection . Comprehensive sampling of C2V3 env sequences from nine HIV-1 infected patients throughout the entire course of HIV-1 infection revealed consistent patterns of viral evolution: both genetic diversity of the viral population at a given time point and mean divergence from the founder strain increased approximately linearly during infection [14] . However , diversity peaked at roughly the same time that viruses using the CXCR4 coreceptors emerged , whilst divergence did not stabilize until close to the time of disease onset . Studies of viral diversity and divergence in HIV-1 patients with different disease progression rates often have conflicting results ( e . g . , [15 , 16] ) . Attempts to distinguish between adaptive and selectively neutral mutations have suggested that slower disease progression is associated with more positively selected sites and higher adaptation rates in env [17 , 18] . From an immunological perspective , however , it remains unclear whether such viral adaptation is the cause or consequence of variability in disease progression rates . Although the physiological processes leading to CD4+ T cell depletion and AIDS are not clearly defined , it is widely accepted that persistent immune activation has a pivotal role in driving disease progression [reviewed in 19] . T cell activation is the strongest predictor of progression to AIDS in HIV-1 infected patients [20 , 21] and can determine the rate and continuity of viral replication [19] . Consequently , immune activation could also impose important constraints on viral generation times and HIV evolution . However , such effects have not been established through evolutionary analyses . The ratio of nonsynonymous/synonymous substitution rates has proved useful in investigating molecular adaptation; however , changes in the absolute rates of nonsynonymous and synonymous substitution should provide greater insight [22] . Changes in synonymous substitution rates can reflect changes in generation time or mutation rate , while nonsynonymous rates can also be affected by changes in selective pressure and effective population size . Previous studies of HIV evolution have typically assumed that the rate of neutral or synonymous change ( per month or year ) is approximately constant among patients ( e . g . , [15 , 16] ) . This assumption may be inappropriate , as the rate depends on the average number of viral generations per unit time , which may vary . For example , viral replication rates can depend on the state of immune activation [23] , viral strains may differ in their replicative ability in different environments [12] , and average virus generation times can be affected by the dynamics of latently infected cells [24 , 25] . To investigate these issues in HIV infection , we developed a new statistical approach to estimate absolute rates of synonymous and nonsynonymous substitutions and to determine how those rates change through time . Our method extends previous relaxed-clock methods with codon model analysis and allows evolutionary rates to change in an uncorrelated fashion along branches in a genealogy [26] . By comparing evolutionary rates for specific branches in within-host HIV phylogenies , we correct for the potentially biasing effects of transient deleterious polymorphism . Using this approach , we investigate the relationships among viral substitution rates , disease progression , and host immune responses . Our results show that disease progression among patients is predicted by synonymous substitution rates , most likely reflecting different levels of persistent immune activation , while nonsynonymous rates evolve within patients as a consequence of changing antibody selective pressure .
We estimated rates of nonsynonymous ( dN ) and synonymous ( dS ) divergence using longitudinal env sequence data from nine HIV-1 infected individuals [14 , 27] . The high mutation rate of HIV is expected to lead to a considerable deleterious mutation load in the viral population , such that the most recent mutations , segregating on external branches of HIV phylogenies , are likely to be deleterious [28 , 29] . This effect is present in each of the nine patients analyzed; mean substitution rates were consistently higher for external branches than for internal branches ( Figure 1 ) . Simulations indicated that such differences were not expected under a neutral model ( Table S1 ) . We also found no evidence supporting a link between high external rates and recombination rates for each time point ( Table S2 ) . Transient deleterious mutations do contribute to HIV evolution but not to the process of nucleotide substitution , since they are , most probably , rapidly purified [29] . Because inclusion of deleterious mutations could bias our estimates of divergence and absolute evolutionary rate , we estimated mean rates of divergence for two sets of branches: internal branches , and “backbone” branches representing the central lineage of the phylogeny that persists through time ( Figure 1; see Methods for a phylogenetic definition of the backbone; specific details of the backbone , internal , and external branches for the other patients can be found in Figure S1 ) . Estimates of synonymous and nonsynonymous divergence for internal and backbone branches are shown in Figure 2 . When patients are classified as moderate or slow progressors ( based on progression time , or the time it takes for the CD4+ T cell count to drop below 200 cells/μl; [17] ) , virus populations in slow progressors appear to accumulate synonymous substitutions more slowly than those in moderate progressors ( Figure 2 ) . This relationship is not apparent for nonsynonymous divergence . We investigated correlations between HIV divergence rates , estimated from the root of the within-host genealogies to progression time , and three continuous parameters that relate to disease progression: progression time , the rate of CD4+ T cell count change over time , and the rate of log viral load ( log VL ) change over time ( Figure 3 ) . The log of the backbone rate of synonymous divergence shows a strong negative correlation with both progression time ( Pearson correlation coefficient r = −0 . 79 , p = 0 . 011 ) and the change in CD4+ T cell count ( r = −0 . 72 , p = 0 . 028 ) , and a moderate positive correlation with the change in log VL ( r = 0 . 65 , p = 0 . 059 ) . No significant correlations were observed for nonsynonymous divergence rates ( r = −0 . 32 , p = 0 . 40 for progression time; r = −0 . 54 , p = 0 . 135 for CD4+ T cell count change; r = −0 . 14 , p = 0 . 58 for log VL change ) . Similar results were obtained when divergence rates were estimated from internal branches ( Figure S2 ) . In contrast to backbone and internal rates , no significant correlations were observed for both dS and dN rates on external branches . Similar results were also obtained when datasets were restricted to samples up to about 70 months after seroconversion ( Figure S3 ) , indicating that the differences in dS rate estimates could not be attributed to differences in time length of sampling . In general , backbone dS rates before progression time seem to show little temporal fluctuation in the trees since dS divergence accumulated in a linear fashion , with R2 values close to 0 . 96 , except for patient 9 ( R2 = 0 . 83 ) . We also investigated the heterogeneity of synonymous and nonsynonymous substitution rates within the env C2V3 gene , because strong site-to-site variation in synonymous rates has the potential to bias dS estimates [30] . Although this analysis revealed very strong site-to-site variation in dS rates , the inferred rate distributions were nearly identical among all patients ( Table S3 ) . Finally , recombination rate estimates ( Table S2 ) did not provide any evidence that recombination could be the cause of the differences between dS estimates . The variability in dS rates could reflect differences in either viral mutation rate or viral generation time , but only the latter provides a likely explanation for the correlation between dS rates and disease progression . However , viral generation time is also expected to affect dN rates to some extent . While we did not observe a significant correlation , this may be because dN rates are determined primarily by the strength of selection on viral mutations , rather than by the absolute rate at which those mutations are generated . Furthermore , most diversifying selection in env results from nAb responses [31] , which are not expected to moderate replication rates and disease progression . To demonstrate this , we further analyzed env sequences sampled through time from two patients with markedly distinct rates of phenotypic escape from nAb responses [31] ( patient 01–0083 and patient 01–0127; Figure 4 ) . As in Frost et al . [2] , we show that the virus that rapidly escaped nAb responses in patient 01–0127 accumulated nonsynonymous substitutions at a considerably higher rate on backbone branches , while synonymous divergence rates appear to be unaffected ( similar plots were obtained for internal branches , unpublished data ) . Because it has been shown that viral divergence stabilizes close to disease onset [14] , we estimated mean divergence rates prior to the progression time in the analyses above . Two hypotheses have been proposed to explain this stabilization: reduced availability of target cells late in infection ( the “cellular exhaustion” hypothesis ) , or reduced selective pressure because of deteriorating immune responses ( the “immune relaxation” hypothesis ) [32] . A recent statistical analysis provided support for the immune relaxation hypothesis by showing that nonsynonymous divergence stabilizes at about the same time as progression time , while synonymous divergence does not [32] . Our analysis provides further evidence that nonsynonymous divergence stabilizes in some patients and that this is less pronounced for synonymous divergence ( Figure 2 ) . Using the empirical relaxed-clock approach , we directly estimated dN and dS substitution rates before and after progression time ( Table 1 ) . These estimates indicate that dN is significantly lower after progression time both for internal and backbone branches ( paired t test: p = 0 . 012 and p = 0 . 001 , respectively; Wilcoxon signed rank test: p = 0 . 016 and p = 0 . 008 , respectively ) , while there is no significant difference in dS before and after progression time ( paired t test: p = 0 . 424 and p = 0 . 333 , respectively; Wilcoxon signed rank test: p = 0 . 461 for internal and backbone branches ) . As an extension to the analysis of closely related HIV-1 strains , we further explored differences among within-patient substitution rates for more divergent HIV lineages . Table 2 lists average dS and dN rates for HIV-1 subtype B infected patients , HIV-1 group O infections , and HIV-2 datasets . Although studies of HIV-1 group O infection are limited , no differences in disease progression between group O and group M infections have been observed [33] . HIV-2 , on the other hand , is known to be less transmissible and less pathogenic than HIV-1 group M [34 , 35] . The two HIV-1 group O datasets had dS rates that were similar to the rates estimated for HIV-1 subtype B moderate progressors ( Table 2 ) . Clinical data for these HIV-1 group O patients indeed indicate progression times in the range of moderate progressors ( 60 and 70 months; [33 , 36] ) . In contrast , HIV-2 patients had dS rates that were more comparable with HIV-1 group M slow progressors ( Table 2 ) . The two published HIV-2 datasets ( P7P8 and P9P10 ) represent cases of perinatal transmission only diagnosed in adulthood [37] , as expected , given the slow nature of HIV-2 disease progression . HIV-2 dN rates are in the same range of HIV-1 dN rates in both moderate and slow progressors . Interestingly , the observation that group O viruses are characterized by a high dN is in agreement with a population-level study that identified more sites under positive selection in env for HIV-1 group O than for HIV-1 group M or HIV-2 [38] .
The relationship between HIV evolution and disease progression is central to our understanding of immune control and to vaccine design . Although much research effort has been focused on this issue , a clear and coherent picture of HIV evolutionary dynamics has yet to be presented . Here , we develop a novel computational approach to estimate the absolute rates of synonymous and nonsynonymous substitution during the course of HIV infection . Our analyses reveal that slower progression to AIDS is strongly correlated with a slower rate of synonymous substitution , indicative of a slower replication rate and longer viral generation times . This reasoning assumes that synonymous substitutions are selectively neutral . Although synonymous mutations may influence fitness in viral populations , experimental exploration of fitness effects caused by single-nucleotide substitutions in vesicular stomatitis virus clearly showed that synonymous changes were roughly neutral [39] . The strong site-to-site variation in dS does suggest that selection is acting on synonymous substitutions in the HIV-1 env gene , but the highly similar distribution of rate variation implies that possible biases , if any , will affect all patients similarly . The similar site-to-site variation in dS for all patients also suggests that slower disease progression is not associated with stronger purifying selection , which could be a less likely alternative to explain the lower dS rates . Because persistent immune activation is an important process underlying disease progression [19] , it may provide a plausible explanation for differences in replication rates . The transcriptional machinery within the nucleus of the host cell is responsible for viral gene expression , making the HIV life cycle critically dependent on the activation state of the host cell [40] . HIV can infect resting CD4+ T cells , but after reverse transcription the pre-integration complex is degraded unless the cell is activated within a few days [41] . Immune activation , which is a strong predictor of progression to AIDS [20] , is therefore a major stimulus for viral replication at the cellular level [40] . In addition , it is believed that T cell activation recruits large numbers of rapidly proliferating , short-lived , activated target CD4+ T cells , thereby creating the ideal condition for increased viral replication , while at the same time causing premature aging and exhaustion of the naïve T cell pool [23] . Experimental evidence indicates that this process is characterized by repeated bursts of viral replication , which have been suggested as a critical driver for the continuity of viral replication [19] . Transmission of free virus will only be efficient when target cells are in close proximity [e . g . , 42] , and infection of new cells is largely restricted to microscopic clusters of T cells in lymphoid tissue [e . g . , 43] , emphasizing the need for recruitment of uninfected activated CD4+ cells through immune activation . Because of both proliferation and activation of target cells , it is not surprising that the state of immune activation will affect HIV generation times and impose important constraints on viral evolution . The cause of immune activation is not clearly established , but it has been suggested that defective virus , which is the primary product of HIV replication , can drive immune activation [44] . If HIV itself plays a central role in immune activation , more replication will reinforce further the state of immune activation . Recent findings show that Nef-mediated T cell receptor ( TCR ) –CD3 down-modulation is capable of protecting against immune activation and activation-induced cell death for most simian immunodeficiency viruses ( SIV ) [45] . This function has been generally conserved in primate lentiviral evolution but lost in the chimpanzee precursor of HIV-1 , SIVcpz [45] . Therefore , these findings seem to resolve the longstanding issue of the nonpathogenic nature of SIV infection . In addition , it provides an explanation for the lower pathogenicity of HIV-2 infections [35] and the generally low HIV-2 synonymous substitution rates observed in this study . Nef-mediated TCR-CD3 down-modulation also correlated with CD4+ T cell depletion in SIVsmm infected sooty mangabeys [45] . A similar phenomenon might therefore be important for differences in pathogenicity in HIV-1 moderate and slow progressing patients . In contrast to an active suppression of immune activation in most SIV/HIV-2 infections , HIV-1 Nef can even increase immune activation [45] , and patients infected with nef-defective HIV-1 have been reported to exhibit a slow progressing or even nonprogressing phenotype [46 , 47] . No doubt other viral and host factors will be important to explain differences between HIV/SIV infections . For example , SIVsmm infections usually exhibit high viral loads , while in asymptomatic HIV-2 infections viral loads are often too low to be detected . Recently , a low replicative capacity of HIV-2 compared with HIV-1 variants has been demonstrated [48] . Producing fewer viral particles per replication cycle will result in lower viral loads and may also constitute a lower stimulus for immune activation . While immune activation provides a likely cause for the observed differences , there may be other explanations for a generation time effect in patients with different disease progression . For example , there may exist an evolutionary tradeoff between immune escape and replication rate . In the absence of immunological pressure , natural selection for fast replication rates would dominate viral evolution [49] . In the presence of an immune response , however , a tradeoff between immune escape and replication rate might be expected . There is evidence that the CTL escape response can indeed reduce replicative fitness by forcing the virus to endure detrimental CTL escape mutations [12 , 13] . Therefore , a vigorous CTL might be indirectly responsible for control of viral replication , longer HIV generation times , and slower disease progression . Differences in replication rate might also reflect differences in the contribution of the latent HIV reservoir to the circulating virus population . The pool of resting memory CD4+ T cells that carry integrated proviral genomes represents a stable reservoir for latent HIV infection . Although they may produce only a fraction of circulating viruses , modelling studies predict that such latently infected cells can still considerably impact mean generation times and replication rates [24 , 25] . The contribution of this reservoir to the free-floating virus population will become more important as CTL killing of infected activated CD4+ T cells becomes more efficient . Latently infected cells are also presumed to play a central role in the continuity of viral replication during chronic immune activation [19] . Between bursts of viral replication induced by immune activation , shedding of the virus in memory CD4+ T cells may be required [19] . Because it is hypothesized that replicative bursts are dependent on initial activation of these memory CD4+ T cells , this might be an additional mechanism by which viral generation times are affected by the state of immune activation . Multiple , overlapping , and nonsynchronized bursts will occur in spatially separated lymphoid tissue [19] , which explains why viral replication may still be continuous and average genetic divergence can increase in an approximately linear fashion . Although CTL escape/viral attenuation and the dynamics of the latent reservoir can be involved in HIV generation times , they represent somewhat less direct explanations and require more experimental validation . Our results show that the nonsynonymous rates in the env gene , most likely effecting phenotypic escape from nAbs ( Figure 4 , [2] ) , do not correlate with disease progression in patients harbouring HIV-1 group M viruses . While these conclusions are based on serial sampled data for only two patients , Frost et al . [2] have clearly demonstrated this effect in multiple patients . Crucially , we show that the selection pressure on env obscures the expected correlation between overall nucleotide substitution rates and disease progression , emphasizing the need to disentangle the contributions of synonymous and nonsynonymous substitutions to HIV evolution . Once these absolute rates are separated , we observe a significant decrease in dN rates after progression time , confirming the hypothesis of immune relaxation during the AIDS stage [32] . We conclude that this immune relaxation reflects attenuated humoral immunity , which is probably a consequence , rather than a cause , of variation in disease progression . This concept is supported by findings that AIDS patients with deteriorated immune responses usually have lower antibody titers than asymptomatic patients ( e . g . , [50] ) . Because it has been suggested that CXCR4-using viruses are subjected to stronger immune control in vivo [51] , immune relaxation also provides an explanation of why these variants with a high in vitro replication capacity and increased target cell range generally do not appear until relatively late in HIV infection . Our results suggest that nAbs are not an important factor in HIV disease progression . In this context , it is interesting to note that autologous nAb response is poor or nonexistent both in HIV-1 elite controllers [52] and HIV-2 infected patients ( N . Taveira , unpublished data ) . However , previous studies have reported higher adaptation rates and more persistently positively selected sites in slow-progressing patients [17 , 18] . On the one hand , this discrepancy may suggest that nAbs , in particular broadly neutralizing antibodies , are able to moderate disease progression to some degree . On the other hand , we cannot exclude the possibility that adaptation rates and positive selected sites in env may reflect CTL selective pressure to some extent . Finally , phylodynamic processes , which explain the net viral adaptation rate as the interaction between viral abundance and selective pressure [53] , must also be considered in HIV evolution . More research is needed to establish the relationship among immune responses , viral adaptation rates , and disease progression in HIV infection . Although a procedure for estimating rates of synonymous and nonsynonymous substitution through time has been developed previously [22] , there are some advantages to this approach . While Seo et al . [22] assume that the tree topology is known without error , we allow for uncertainty in the reconstructed genealogy . Our definition of internal and backbone branches is a phylogenetic one , but we average over a set of plausible trees in an empirical fashion and we also incorporate sampling time information . Seo et al . [22] were able to test whether changes in synonymous and nonsynonymous rates are correlated , whereas we model changes in the overall substitution rate ( in an uncorrelated fashion [26] ) , after which the synonymous and nonsynonymous component for each branch is decoded using maximum likelihood methods . It is possible that a full probabilistic treatment of genealogical uncertainty and independent changes in synonymous and nonsynonymous rate could be developed . However , overcoming the technical hurdles required to achieve this objective is beyond the scope of this study . In this genealogical framework , we do not account for the process of within-host recombination , which can considerably shape HIV diversity [54] . Although recombination could lead to inflated dN/dS estimates [55] , there is no reason to expect that recombination would affect synonymous rates in a way that correlates with disease progression . This was supported by the lack of clear correlations between dS and recombination rate estimates . In conclusion , our evolutionary analysis revealed a strong correlation between synonymous substitution rates and HIV disease progression . Further work will be necessary , however , to clearly establish the mechanisms that determine HIV generation times . A combined approach incorporating both experimental and computational techniques should provide important insights into the patterns of HIV evolution and disease progression .
Our approach to infer synonymous and nonsynonymous substitution rates , and to explore how these rates change through time , is an empirical extension of recently developed Bayesian relaxed-clock models [26] . Given a fixed tree topology , branch lengths measured in units of expected synonymous and nonsynonymous substitutions can be estimated using codon substitution models . We applied a codon-based extension of the Hasegawa–Kishino–Yano model ( HKY85 ) ( MG94xHKY85 with codon equilibrium frequencies estimated from position-specific nucleotide frequencies , [56] ) , for which the rate matrix for substituting codon x with codon y in infinitesimal time , Qx , y ( α , β , κ ) , is given by: In this parameterization , α denotes the synonymous substitution rate , while β denotes the nonsynonymous substitution rate , and their ratio ( ω = β/α ) reflects the strength of selective pressure along a specific branch . κ is the transversion/transition ratio , and πny represents the frequency of the target nucleotide at the appropriate codon position . The parameters α and β can be shared among branches ( equivalent to the “one-ratio” model; [57] ) or allowed to take up branch-specific values ( equivalent to the “free-ratio” model; [57] ) . We applied the latter , dubbed the “local” codon model , which has the following form for the expected number of substitutions per site on branch bi: where f1 , f2 , g1 , and g2 are functions determined by the nucleotide composition of the sequence alignment and shared among branches [58] . The first term in the sum corresponds to the contribution of synonymous substitutions ( a product of the branch-specific α ) and the second to the contribution of nonsynonymous substitutions ( a product of the branch-specific β ) . The time parameter t is not estimable alone but products tαi and tβi are . In this codon model , the time parameter t could be estimated assuming a dated tip molecular clock model , thereby providing absolute rates of synonymous and nonsynonymous substitution [59] . To estimate changes in synonymous and nonsynonymous rates , however , the assumption of a strict molecular clock needs to be relaxed ( e . g . , [22] ) . Recently , a relaxed-clock approach has been developed that is applicable to measurably evolving populations and takes into account genealogical uncertainty by averaging over a set of plausible trees [26] . Since the sampling of genealogies using codon models is too computationally expensive at present , we applied an empirical extension of the relaxed phylogenetic model and sample trees under the equivalent nucleotide substitution model . We used Markov chain Monte Carlo ( MCMC ) methods as implemented in BEAST 1 . 3 to obtain a posterior distribution of trees under an uncorrelated relaxed clock [60 , 61] . In this approach , the nucleotide substitution rate on each branch of the tree is drawn independently and identically from an underlying rate distribution [26] , in this case an exponential prior distribution among branches . Our full Bayesian probabilistic treatment is characterized by the following posterior distribution: where g represents the tree topology and Θ contains the hyperparameters of the tree prior . We used a piecewise-constant model of population size , the Bayesian skyline plot model , that provides population size estimates for each coalescent interval of the genealogy g [62] . Θ contains both the parameters for the group sizes , which define the number of coalescent events in each grouped interval , and the population size for each group . The relaxed-clock parameter λ represents the exponential distribution of rates across lineages ( with mean and standard deviation λ−1 ) [26] . Because we used the HKY85 of nucleotide substitution with gamma-distributed rate variation among sites , the vector Ω contains the transition/transversion ratio ( κ ) and the shape of the gamma distribution ( α ) . The posterior density was investigated using MCMC with the length of the chains , sampling frequency , and burn-in dependent on the dataset analyzed . The MCMC samples were inspected for convergence to stationarity , and effective sampling sizes were calculated using Tracer 1 . 2 [63] . Using the relaxed-clock approach , a posterior distribution of tree topologies can be obtained with branch lengths in units of time and in units of the expected number of nucleotide substitution per site , both related by the vector of rates for the branches [26] . To infer branch lengths in the expected number of synonymous and nonsynonymous substitutions separately , we applied the local codon model to a set of trees sampled from the posterior distribution . To achieve computational tractability and convergence under this parameter-rich scheme ( for reasonably large datasets , both in terms of the number of taxa and the number of posterior trees ) , we inferred the maximum likelihood estimates of the parameters while constraining the branch lengths ( Esub ) to the codon-rescaled estimates obtained by the Bayesian relaxed-clock analysis . This allowed us to determine the nonsynonymous and synonymous component of the branch lengths , and using the Bayesian estimate for branches in time units for each tree , absolute rates of nonsynonymous and synonymous substitutions could be inferred and empirically averaged over a set of genealogies . It has been shown that fixing branch lengths has a negligible effect on secondary nonsynonymous and synonymous rate inference [64] . For each dataset , maximum likelihood estimation was performed using HYPHY version 0 . 99 based on 200 trees sampled from the posterior distribution [65] . Scripts to perform local codon analysis on a set of trees in HYPHY are available at http://evolve . zoo . ox . ac . uk and http://www . hyphy . org . Within-gene variation of synonymous and nonsynonymous substitution rates was analyzed using codon models that incorporate site-to-site heterogeneity of both dS and dN [30] . Under the uncorrelated relaxed clock , it is possible to infer individual rates for all branches in a genealogy and , thus , also a mean rate for each subset of branches . Our analysis focuses on within-host HIV genealogies , for which terminal branches may have an excess of short-lived deleterious mutations that have not reached fixation . Therefore , we estimated mean substitution rates for internal branches and for the central trunk of the ladder-like genealogies , referred to as the “backbone” ( Figure 1 ) . We define the backbone as the set of branches connecting the root of the tree to the sequences sampled at the last time point , excluding both terminal branches and internal branches from a common ancestor of sequences sampled at the last time point only . Weighted averages were used to report mean substitution rates for a set of branches , as defined by Drummond et al . [26] . Mean substitution rates before or after progression time were calculated as weighted averages for the set of branches ( or partial branches ) before or after that time point in the tree . Nonsynonymous and synonymous divergence over time was estimated by calculating the mean accumulation of substitutions along the lineages in a particular time interval and averaging them over the set of genealogies . Software to plot divergence over time and to obtain substitution rates for a subset of branches is available from the authors on request . Substitution rates in relationship to disease progression were investigated in nine patients extensively sampled over a 6–13 . 7 year period starting close to the time of seroconversion [14 , 27] . Both the Shankarappa et al . [14] C2V3 env sequence data and the Shriner et al . ( 2004 ) [27] follow-up sequences were analyzed . Complete env gene sequences longitudinally sampled from two patients with distinct phenotypic escape form nAbs were obtained from Frost et al . [2] . Heterochronous HIV-1 group O env data were obtained from the two available intrapatient evolution studies [33 , 36] . For HIV-2 , we analyzed env clones from two serially sampled patients [66]; the sequence “C9/1997” from patient C was excluded from the analysis because it has been identified previously as a recombinant [67] . In addition , we inferred substitution rates from population sequences sampled at different time points in four separate patients [68] . Since only three or four sequences were available per patient , these sequences were analyzed as one single dataset with the internal and backbone rate calculated as the mean rate for only within-host internal branches and backbone branches , respectively . Finally , we analyzed data from three HIV-2 mother–child transmission pairs [37] , including unpublished sequences for one pair ( P3P4 ) . These sequences were obtained using methods described elsewhere [37] and are available from GenBank . Since only one sample of cloned sequences was available for both mother and child at the same time , rates were estimated by specifying a normal prior distribution on the time to the most recent common ancestor based on the time of birth; monophyletic constraints were imposed on both the mother and child sequences . HIV-1 group M and HIV-2 alignments were trimmed so that approximately the same gene region was investigated for all datasets ( position 823 to 1128 relative to HXB2 env and position 823 to 1134 relative to SMM239 env ) .
The GenBank ( http://www . ncbi . nlm . nih . gov/GenBank ) accession numbers of the HIV-2 sequences discussed in this paper are DQ787116–DQ787121 .
|
During the clinical course of HIV infection , an asymptomatic phase always precedes the acquired immunodeficiency syndrome ( AIDS ) . The duration of this asymptomatic phase is highly variable among patients and reflects the rate at which the immune system gradually deteriorates . Although humoral and cell-mediated immune responses are mounted against HIV , continuous replication and adaptation allows the virus to escape host immune responses . To gain a better understanding of the role of viral evolution in disease progression , we developed a new computational technique that can estimate changes in the absolute rates of synonymous and nonsynonymous divergence through time from molecular sequences . Using this type of evolutionary inference , we have identified a previously unknown association between the “silent” evolutionary rate of HIV and the rate of disease progression in infected individuals . This finding demonstrates that cellular immune processes , which are already known to determine HIV pathogenesis , also determine viral replication rates and therefore impose important constraints on HIV evolution .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"mathematics",
"computational",
"biology",
"genetics",
"and",
"genomics",
"nematodes"
] |
2007
|
Synonymous Substitution Rates Predict HIV Disease Progression as a Result of Underlying Replication Dynamics
|
Neural networks , despite their highly interconnected nature , exhibit distinctly localized and gated activation . Modularity , a distinctive feature of neural networks , has been recently proposed as an important parameter determining the manner by which networks support activity propagation . Here we use an engineered biological model , consisting of engineered rat cortical neurons , to study the role of modular topology in gating the activity between cell populations . We show that pairs of connected modules support conditional propagation ( transmitting stronger bursts with higher probability ) , long delays and propagation asymmetry . Moreover , large modular networks manifest diverse patterns of both local and global activation . Blocking inhibition decreased activity diversity and replaced it with highly consistent transmission patterns . By independently controlling modularity and disinhibition , experimentally and in a model , we pose that modular topology is an important parameter affecting activation localization and is instrumental for population-level gating by disinhibition .
Activity gating and control over propagation are fundamental capacities of neural circuits . It is widely accepted that population-level gating is strongly affected by changing the balance between excitation and inhibition in connected sub-populations of neurons [1–6] . However , while the role of excitation-inhibition has been widely investigated , the contribution of circuit topology to activity gating has received much less attention . Modular topology is of particular interest , as it is a fundamental feature of biological neuronal circuits [7–11] . Modular circuits are composed of highly connected groups of neurons ( modules ) which are loosely connected to other groups . Such a network organization is found at many spatial scales , ranging from anatomically defined brain regions to groups of neurons [7 , 8 , 12 , 13] . Experimental investigation into the contribution of modular topology to gating phenomena is faced with major challenges . While large brain areas have well-documented connectivity and activity maps [12 , 14 , 15] , accessing complete brain circuits at smaller scales is still limited . Foremost , the connectivity maps are highly untraceable in the three-dimensional architecture of the tissue . In addition , since cell assemblies are often dispersed in space , their simultaneous identification and recording are still beyond the reach of contemporary technologies [16] . Finally , as neuronal circuits are not prone to design , systematic studies are impossible . Consequently , studies aimed to relate activity propagation and gating to network architecture ( such as modular networks ) are mainly restricted to theoretical investigations [17] . Indeed , theoretical studies indicate that modular organization greatly impacts network functionality [18] . Modular circuits provide control over activity propagation [9 , 10 , 19 , 20] , time-scale separation [21] , dynamical complexity and the computational capacity of the network [10 , 22–25] . In this study we experimentally address , for the first time , the relation between circuit modularity and activity gating . To overcome the inherent limitations associated with studying intact tissues , we utilized a cell patterning technique to induce self-organization of modular networks in culture . We found that pairs of connected modules support conditional propagation that is dependent on the activity intensity in the sending module . In large networks of many connected modules , conditional propagation enhances the diversity of activation patterns , and is manifested as events initiated at different modules which then propagate to different distances . Interestingly , blocking network inhibition decreased the activity diversity and replaced the conditional propagation with highly reliable transmission . These features are absent in the activity repertoire of uniform networks . Thus , we show how a combination of modular circuit architecture and disinhibition supports gating .
To investigate how activity propagates through modular neuronal networks , we used a unique biological model system of engineered clusters in vitro . Control over circuit architecture was achieved using heterogeneous surfaces with different degrees of adhesiveness ( see Materials and Methods ) . Specifically , islands of highly adhesive surfaces were realized on a non-adhesive background . Due to their innate propensity to cluster , neurons self-organized into modular circuits within several days in culture ( Fig 1 ) , in accordance with previous work [26–28] . Each of the clusters comprised of several tens to hundreds of neurons ( S1A Fig ) , connected through a bundle of fasciculated neurites ( Fig 1 ) . The number of cells per cluster was estimated from the cluster area using N = 0 . 0079S-1 . 9 , where N is the number of cells and S is the cluster area in μm2 . This relation was calculated in a previous publication in which clusters were grown under the same experimental conditions [29] . In contrast to previous methods [30–33] , our procedure allows the formation of small-scale sub-networks of different sizes , ranging from cluster chains ( Fig 1B ) to two-dimensional networks of many connected clusters ( Fig 1C ) . The position of each cluster was aligned with a micro electrode , allowing local electrical recording from each cluster ( Fig 1B and 1C-right ) . Using the embedded electrodes ( Fig 2A ) , the simultaneous activity of all clusters was recorded . The vast majority of measured clusters were found to be spontaneously active after several days in culture . A typical voltage trace recorded from one electrode is shown in Fig 2B . Each electrode recorded the superimposed activity of many neurons within each cluster . It was previously shown that voltage traces represent spike summation and correspond to the increase and decrease in population activity [29 , 34] . Accordingly , to represent the activity intensity of such clusters , we averaged the rectified voltage traces over short time windows and color coded them ( Fig 2C ) ( see Materials and Methods for details ) . Spontaneous activity of individual clusters was characterized by typical features previously observed in developing networks [29 , 35 , 36] . We observed the activation of network bursts ( NBs ) which are short epochs of network intense firing , separated by longer periods of sporadic activity ( Fig 2D ) . To verify that these events represent the collective activity of many neurons that synchronize within the NB time window , we performed recordings from single clusters using dense electrode arrays ( 30 μm spacing ) with a smaller electrode surface area ( 314 μm2 in contrast to 2827 μm2 in our regular electrodes ) which pick up activity from more local populations of neurons . The variability in the spiking profiles across different electrodes exemplified the synchronized nature of spiking within NBs of single clusters ( S2 Fig ) . While the activity within clusters appeared to be synchronized , between connected clusters the synchrony was transient . We begin by examining the activity of connected cluster pairs ( Fig 3A ) . NBs were found to be either confined to a single cluster or spread over nearby connecting clusters ( Fig 3B ) . To investigate whether the activity propagation between connected clusters was related to the activity intensity in the clusters , we examined the propagation between pairs of adjacent clusters . For each pair , we defined one cluster as the "sending cluster" and detected all NBs occurring in this cluster ( see details in Materials and Methods ) . From this NB pool , we selected only NBs which were activated in the sending cluster before the neighboring cluster , defined here as the "receiving cluster" . This selection process consisted of rejecting NBs according to the delay calculated from the peak position of the cross-correlation function of the smoothed ( convoluted with a Gaussian kernel , σ = 10ms ) activity intensities of the sending and receiving clusters . Positive offset delay was attributed to propagation from a receiving to a sending cluster . Fig 3A–3C illustrates activity propagation in a representative cluster pair . 300 consecutive NB traces , recorded from cluster 1 ( the sending cluster ) in Fig 3A , are shown in Fig 3C-left . These NBs were reordered according to increased NB intensity ( sum over the NB activity intensity trace ) in the sending cluster . The receiving cluster responses to these NBs are shown in Fig 3C-right . The total intensity of these responses ( averaged over all NBs ) were only slightly lower than in the sending cluster ( Z-score of the difference in AI was 0 . 083 ) , but significant ( PV = 0 . 013 , Mann-Whitney-Wilcoxon test ) . Low intensity NBs did not propagate to the receiving cluster , while strong NBs did . It should be noted , however , that a small fraction of strong NBs failed to propagate . We termed this selective activation of NBs in the receiving cluster following an NB in the sending cluster as conditional propagation . To quantify this behavior over many cluster pairs , we calculated for each NB in the selected pool , and for each cluster pair , the probability of having different normalized NB intensities ( intensities divided by the intensity standard deviation ) in the receiving cluster as a function of normalized NB intensities in the sending cluster . Data of the cluster pair in Fig 3A are presented in Fig 3D-left . This representation further illustrates that low intensity NBs did not yield strong responses in the receiving cluster , while NBs stronger than a certain value successfully propagated to the receiving cluster . It is important to note that in the example shown here , a distinct threshold between non-propagating and propagating NBs is apparent ( Fig 3D-left ) . To quantify this threshold-like behavior , we calculated the bi-modality measure [37] on the distribution of normalized NB intensities projected on the identity line ( S4A Fig ) for the cluster pair in Fig 3D-left ( S4B Fig ) , and for all cluster pairs ( S4C Fig ) . Most clusters showed values larger than ( 0 . 555 ) corresponding to a tendency to bi-modality over uni-modality ( S4C Fig ) . We verified these results by performing visual inspection of the distributions and determined that nearly half ( 48% ) of the pairs showed a clear intensity threshold in the propagation probability . However , taking into account also pairs that did not show strong bi-modality , the general rule was that strong activations in the sending cluster yielded strong responses in the receiving cluster and vice versa . This effect was quantified by calculating the correlation between the normalized NB intensity of the receiving and sending clusters for all NBs in 163 cluster pairs , from 26 cultures ( Fig 3D-right ) . Two types of cluster networks were considered in the analysis: The first type is one dimensional chain of clusters and the second involves a cluster chain with one of the clusters connected to a larger network ( of many connected clusters ) . Within these networks , only pairs connected exclusively through a neurite bundle ( and not through any other pathway ) were analyzed . The data in Fig 3D-right exhibits a clear preference towards positive values , further suggesting that information about firing intensity is utilized by neuronal networks to control propagation between connected sub-populations . No significant correlation was observed between NB intensity correlation strength and the normalized difference in cluster cell numbers ( C = 0 . 084 , PV = 0 . 28 ) . We next sought to investigate how modularity affects propagation delays . Delays in our engineered networks were evaluated by extracting the location of the peak in the cross-correlation function of every pair and every NB ( as previously described ) . Only NBs propagating from the sending cluster to the receiving cluster were considered . The delay distribution across all NBs is shown in Fig 3E-left . The average delay was several tens of milliseconds ( red line in Fig 3E-left ) and the maximal value reached was 250 ms . With a distance between clusters of 500 μm , a delay of 100 ms corresponds to a propagation speed of 5 μm/ms . Long delays were also observed in other cluster pairs , as shown by the distribution of average delays ( Fig 3E-right ) . We did not find a significant correlation ( C = 0 . 060 , PV = 0 . 43 ) between the average delay and the normalized absolute difference in cell numbers ( |N1−N2|2 ( N1+N2 ) , Ni being the number of cells in each cluster ) . Interestingly , a weak positive correlation was found between average delay and the number of cells in the receiving cluster ( C = 0 . 202 , PV = 0 . 007 ) ( S1C Fig ) , but not in the sending cluster ( C = 0 . 098 , PV = 0 . 19 ) , suggesting that the delay is associated with the network recruitment time in the receiving cluster . Correspondingly , the average recruitment time for all NBs in a cluster , measured as the time between NB onset and NB peak ( see NB detection in Materials and Methods ) , had similar time scales to the observed average delays ( S3 Fig ) . Fig 4A shows an example of 100 consecutive NBs recorded from a three cluster chain ( only NBs in which all three clusters were active are shown ) . Visual inspection clarified that most of the time cluster 2 fired before cluster 3 , indicating that propagation in cluster chains is asymmetric . To examine the different propagation patterns and their abundance in this network , we clustered them by calculating the similarity matrix between NBs using a dendrogram [38] ( see Materials and Methods ) ( Fig 4B ) . For clarity , only NBs in which at least two clusters were active are presented . Interestingly , clear NB groups with high similarity are observed ( marked by solid line rectangles in Fig 4B ) . These different groups correspond to different propagation patterns as indicated by the average NB profile within each group ( Fig 4C ) . Propagation in this modular network , although bi-directional , is rarely symmetric . The non-uniformity of the similarity matrix within each rectangle represents NB pattern variation within each group , as seen by examining the coefficient of variation of off-diagonal terms within a group ( 0 . 21 , 0 . 04 , 0 . 21 , 0 . 03 and 0 . 07 in groups 1 to 5 respectively ) . The higher coefficient of variation in groups involving the activation of cluster 1 suggests that specific links in the network are more variable than others . For example , groups 2A and 2B ( see Fig 4C ) are characterized by the same propagation directions , but with a different average delay ( between clusters 1 and 2 ) . The activation probability of different propagation patterns vary considerably ( P = 0 . 137 , 0 . 673 , 0 . 103 , 0 . 005 , 0 . 082 for groups 1 to 5 respectively ) . Such variability can be represented by the pattern entropy ( E = −∑iPilog2Pi = 1 . 006 for this example ) as discussed in detail in the next section . Clustering into propagation groups existed in all network examined ( 69 chains form 26 cultures ) . However , this grouping was highly variable between networks both in terms of separability between groups and the number of groups . Consequently , to reliably quantify asymmetry , we needed a measure that reduces the complexity of patterns to simple asymmetric relations between pairs . To do so , we used the cross-correlation function between connected clusters over long recordings of eight hours ( Fig 4D ) . We subtracted the integral over the positive side of the cross-correlation function from its negative side and divided it by the total sum . We analyzed only positive correlations between clusters , thus negative cross-correlation values were not included in the integral . The resulting long-term asymmetry is a measure between -1 and 1 with the sign determining the propagation direction and the value corresponding to the asymmetry magnitude . For example , in Fig 4A–4C , propagation pattern 2 dominated the network's activity ( Fig 4B ) . This pattern represents events initiated in cluster #2 and propagated to the neighboring clusters . Correspondingly , the asymmetry measure for cluster 1 and 2 is -0 . 21 and for cluster 2 and 3 is 0 . 31 . Pooling over all cluster pairs ( n = 89 ) we find that the average absolute asymmetry was 0 . 26±0 . 21 ( mean ± standard deviation ) . Such asymmetry is in accordance with previous reports indicating that activity asymmetry exists between connected sub-populations , even if they are very similar to each other [32] . We next examined if asymmetry is affected by the relative differences in cell number between connected clusters . Although the differences between cell numbers were not large ( S1B Fig ) , we found a small positive correlation between the normalized cell count and the asymmetry value ( C = 0 . 331 , PV = 0 . 001 ) ( S1D Fig ) . An interesting question that arises from these results is whether and how activity asymmetry is affected by asymmetry in the cluster pair , and by the manner in which this cluster pair is connected to the rest of the network . To address this question , we applied the asymmetry measure to pairs of clusters in which only one of the clusters was connected to a larger clustered network . Such pairs had a clear structural asymmetry . A schematic drawing of such pairs is shown in the inset of Fig 4E , where c1 and c2 are the two clusters and x denotes a group of connected clusters . We differentiated between two connectivity patterns according to the number of links ( connections to other clusters ) , n , cluster c2 had . The asymmetry statistics for n = 1 and n>1 are shown in white bars and black bars respectively in Fig 4E . In most networks , the long-term activity asymmetry was negative ( Fig 4E ) , corresponding to activity propagation from the cluster group towards the chain's end ( x towards c1 in the schematic drawing in Fig 4E ) . This result suggests that the structural asymmetry of the embedding network contributed to the functional asymmetry of the cluster pair . Thus , if a cluster in a pair is connected to other clusters , it will more likely drive activity in this pair . Interestingly , the average asymmetry of pairs connected through more than one link was more negative than that of pairs connected through one link ( -0 . 26 and -0 . 10 respectively , PV = 0 . 07 , t-test ) , suggesting that the drive is stronger when the structural asymmetry is stronger . We note that asymmetry was calculated on long time series to represent a gross averaged estimation in a valid manner . When inspecting the delay of the cross-correlation function ( or the asymmetry ) on a single cluster pair , we find that asymmetry is modulated in the network ( Fig 4F ) . This is observed in the ratio between propagation to one and to the opposite direction calculated on one hour windows ( black line in Fig 4F ) . Despite this dynamic change , the average values of asymmetry were mostly negative ( Fig 4E ) . We further examined whether we can gate activity propagation by global disinhibition . Specifically , whether we can dramatically increase the probability that NB propagates between clusters by applying inhibitory synaptic blockers ( Fig 5 ) . Under control conditions ( normal growth media ) , cluster chains exhibited a large spectrum of activation profiles from confined bursts ( in single clusters ) to network-wide activation ( Fig 5A ) . Upon disinhibition , using a GABAA ( γ-aminobutyric acid ) channel antagonist ( Bicuculline , 30μM ) , the conditional propagation was replaced by network-wide synchrony ( Fig 5B ) . Careful examination of the propagation patterns within the NBs under disinhibition revealed that not only did the network synchronize to operate as a single unit ( Fig 5B ) , but also the synchrony was characterized by highly ordered propagation patterns ( Fig 5D ) . To monitor these patterns over consecutive NBs , we represented each NB as a vector ( as shown in Fig 5C and 5D-right ) and plotted it , for consecutive NBs , under control conditions ( Fig 5E ) and under disinhibition ( Fig 5F ) . In Fig 5E and 5F , blue dots correspond to cluster activation during an NB , and the arrows correspond to the activity propagation direction ( extracted from the peak lag of the cross correlation function ) . Under disinhibition , the network’s activity collapsed to a stereotypic pattern in which the entire network was fully activated and cluster number 5 functioned as an activation focus . Under disinhibition , 95% of the NBs ( Ntotal = 4097 ) were initiated by cluster 5 , in contrast to only 4% ( Ntotal = 7206 ) under control conditions . Such a repeated propagation pattern is in contrast with the much wider repertoire observed under control conditions ( Fig 5E ) . The variability in NB patterns was further quantified over long-term recordings by defining the NB pattern entropy . Each NB was reduced to a binary series of zeros and ones corresponding to the activation of different clusters ( blue dots in Fig 5E and 5F ) . The occurrence probability for every binary pattern , Pi , was calculated and presented as a pie chart for control and disinhibition conditions ( Fig 5G ) . The entropy , E , was calculated from these probabilities as E = −∑iPilog2Pi . Under control conditions , a wide NB pattern distribution and relatively high entropy value ( E = 2 . 32 ) were observed . However , after disinhibition , the number of patterns dramatically decreased , corresponding to a lower entropy ( E = 0 . 94 ) . For entropy calculations , we used only long enough chains ( three or more active clusters ) that can support a high enough variability in NB patterns . Since every chain was of a different length and consequently had different potential entropy values , the entropy was normalized to the maximal possible entropy in the chain , Emax = log2 ( 2N − 1 ) ( the term -1 was added to subtract the case in which no cluster was activated ) . Five out of six chains exhibited a decrease in normalized entropy following disinhibition ( Fig 5H ) . This transition from a wide to a narrow pattern distribution following disinhibition suggests that under control conditions , networks maintain a certain relation between inhibition and excitation . This relation , in combination with the modular architecture , allowed each cluster to be activated autonomously while still being connected to other clusters , and thus also having the potential to activate them . When disinhibited , the network lost its diversity and collapsed into stereotypic global activation . Disinhibition ( implemented here using Bicuculline ) served as a gating mechanism , altering signal propagation . As illustrated above , modularity introduces new features to the activity repertoire of uniform networks . To explore whether these features are preserved in networks of many connected clusters , we examined large , two-dimensional networks of connected clusters ( Fig 1C ) and contrasted their activity with that of large uniform networks ( Fig 1A ) . The most conspicuous difference appeared to be the degree of synchrony . While uniform networks mostly showed network-wide activation that spanned a large fraction of the cell population ( Fig 6A-bottom; S5 Fig ) , clustered networks exhibited NBs of different sizes that propagated to different distances ( Fig 6A-top; S5 Fig ) . This diversity is the direct result of the conditional propagation inherent in the modular bridge between clusters . In large modular networks , the cumulative number of such bridges between any two clusters increased with the distance between them . We next examined whether co-activation of clusters depended on distance . This effect was quantified by measuring the average Pearson correlation between the activity of cluster pairs for long time periods ( >8 hours ) . A decrease in correlation with distance was evident ( Fig 6C-top ) , highlighting the locality of the activation in large modular networks . Neighboring clusters had a higher probability to fire in synchrony . The tendency for local activation was found to coexist with epochs of global activation ( Fig 6A and 6B-top ) exemplifying the network's potential for activation diversity . Large clustered networks were also characterized by long delays . Delays between cluster pairs were calculated from the peak of the smoothed cross-correlation function and averaged over all NBs ( see Fig 3 ) . These delays accumulated during burst propagation , and in many cases the last cluster to be activated during an NB began its firing long after the first cluster already ceased bursting ( Fig 6B-top ) . Delays ranged from tens to hundreds of milliseconds ( corresponding to an average propagation speed of 6 . 5±0 . 2 μm/ms , mean±SE ) , and increased with the number of bridges between clusters ( Fig 6C-middle ) . Time delays partially affected the decrease in correlation with distance ( Fig 6C-top ) . Conditional propagation was also quantified by calculating the transfer probability . Namely , given the occurrence of an NB in one cluster , the probability that an NB occurred within one second in the other cluster . This probability was also found to decrease with distance ( Fig 6C-bottom ) . Finally , we note that conditional propagation , long delays and high diversity in the network degree of activation and synchrony are much less pronounced in the activity repertoire of large uniform networks . Uniform networks are characterized by large scale network events ( S5 Fig ) . Once a network event is initiated , it quickly propagates ( with an average propagation speed of 22 . 1±1 . 2 μm/ms , mean±SE ) and recruits most of the network ( Fig 6A , 6B-bottom; S5 Fig ) . Uniform topology appears to lead to uniform activation which does not vary much with distance ( Fig 6C ) . Previous studies showed that uniform networks support a mode of partial network activation called aborted bursts , during which only a subset of the population is active [39] . Owing to these aborted bursts ( which are also observed in our data ) the average transfer probability in uniform networks is well below unity ( Fig 6C-bottom ) . However , in uniform networks such aborted bursts are not confined to a local area in the network and are always activated in the same sub-population of neurons [39] . Consequently , neither the transfer probability nor the delays in uniform networks depend dramatically on distance ( Fig 6C-middle ) . As shown above , disinhibition drastically increased propagation between the modular units and induced global network synchronization with defined propagation patterns . In large clustered networks , clusters are connected through multiple pathways which may lead to different disinhibition effects . To test the propagation patterns in large clustered networks before and after disinhibition , we examined the cross-correlations function between clusters over single NBs . Since these are two-dimensional networks ( unlike the one-dimensional cluster chains discussed above ) , we extended our propagation analysis by calculating the propagation vector for every cluster . We first identified NB windows in the network ( see Materials and Methods ) , and calculated the cross-correlations of smoothed ( convolution with a Gaussian , σ = 50ms ) activity traces between every cluster and all of its neighbors , corresponding to the fact that long range connections between the clusters were rare due to the grid-like organization of the clusters . Clusters with very weak activity during the NBs ( active for less than 10 ms ) were not analyzed and only NBs with at least five active clusters were considered . The cross-correlation between every cluster pair was represented by a vector with a magnitude corresponding to the location of the peak of the cross-correlation function , and a direction determined by the physical direction between the clusters ( Fig 7A—gray arrows ) . The propagation vector was calculated by averaging the cross-correlation vectors over all active clusters during each NB ( Fig 7A—blue arrow ) . The angle of this vector with the positive x-axis was marked by ϴ ( Fig 7A ) . The propagation vector for each cluster during 20 consecutive NBs is plotted in the physical space of the network in Fig 7B ( magnitude was set to be the same for all clusters ) . In agreement with the high diversity of activation patterns in clustered networks , different NBs propagated to different directions . This was also evident by examining a larger pool of NBs ( Fig 7D-left ) . Here the angle of the propagation vector , ϴ , was presented in color code for every cluster during consecutive NBs . As in the case of the cluster chains , following disinhibition ( application of 30 μM Bicuculline ) , the high NB pattern diversity was replaced by stereotypic patterns . This shift ( or gating ) in activation profile is represented by the narrow distribution of propagation directions for different clusters over consecutive NBs ( Fig 7C and 7D-right ) . Furthermore , the propagation patterns revealed the emergence of a clear activity initiation focus , similar to the case of the cluster chains discussed above ( Fig 5E and 5F ) . Under disinhibition , 42% of NBs ( Ntotal = 366 ) were initiated in the lower left cluster ( marked by a white dot in Fig 7C ) , in contrast to only 0 . 4% ( Ntotal = 2345 ) under control conditions . The propagation variability of the network was quantified by calculating the standard deviation of ϴ for all NBs in the recording , followed by averaging over all clusters . Under control conditions the propagation variability was 1 . 367 radians , and after disinhibition it decreased to 0 . 775 radians . The mean propagation variability for different networks is presented in Fig 7E-left . In all analyzed networks ( six out of six ) a similar reduction in angle distribution was observed . Reduced variability was also observed in the number of participating clusters during NBs . While in control conditions , the NB sizes varied considerably from activation of single clusters to activation of the whole network ( Fig 7F-right ) ; after disinhibition , most of the NBs were synchronized over the entire network ( Fig 7G-right ) . The variability in cluster activation was quantified by calculating the entropy of cluster activation patterns , as previously described for cluster chains . All analyzed networks ( six out of six ) showed a decrease in activation entropy following disinhibition ( 7E-right ) . In Fig 7F and 7G , we explored the number of participating clusters in NBs with different intensities . We plotted the number of active clusters as a function of NB intensity for control ( Fig 7F-left ) and disinhibition ( Fig 7G-left ) conditions . In these plots , the intensity of each NB was normalized to the number of active clusters during this NB . Both in control and disinhibition conditions , the total normalized network intensity increased with the number of participating clusters . Thus , the extent of global activation ( the number of clusters recruited during the NB ) affected the local degree of activation ( NB intensity in single clusters ) . However , under disinhibition most NBs recruited the entire network , while in control conditions , different NBs recruited a different number of clusters ( Fig 7F and 7G-left ) . This is further illustrated by plotting the distribution of the number of recruited clusters over all NBs in control ( Fig 7F-right ) and disinhibition ( Fig 7G-right ) conditions . The results described above demonstrate that modular topology support gating by disinhibition . To better understand this effect , we developed a computational model based on two coupled clusters in which different network parameters could be systematically modified . As cortical cultured networks exhibit complex organization of dynamical patterns , we adopted a previously published model that reproduced the main features of these patterns [32 , 40] . Neurons ( N = 50 in each cluster ) were modeled as Morris-Lecar elements with modified Tsodyks-Markram synapses and synaptic noise ( see Materials and Methods ) . One out of every five neurons is an inhibitory neuron . For isolated clusters , the connectivity probability within clusters ( intra-connectivity ) was 0 . 25 and 0 . 2 for clusters 1 and 2 respectively , and the connectivity between clusters ( inter-cluster connectivity ) was initially set to zero . Intra-cluster connections were then replaced with inter-cluster connections with a probability λ which is defined as the modularity of the system . At the limit of λ = 0 . 5 the two clusters converge to a large uniform network . The model parameters ( see Materials and Methods ) were chosen to fit experimental data . When the inter-connectivity was set to zero ( isolated clusters ) , expectedly , each cluster exhibited short epochs of network bursts that were separated by sporadic single neuron activation , similar to activity patterns of isolated clusters in culture [29] . For slightly larger inter-cluster connectivity ( λ = 0 . 02 ) , some NBs successfully recruited the connected cluster , while others failed to elicit an NB in the connected cluster ( Fig 8A ) . To quantify this property , we calculated the transfer probability for different modularity values . To do so , we detected NBs in the two clusters . An NB was considered as transmitted if following the activation of cluster 1 , an NB peak was detected in cluster 2 within a time frame of 200 ms ( see Materials and Methods ) . Since both clusters were spontaneously active , the transfer probability was non-zero even if the clusters were disconnected ( λ = 0 ) . To compensate for this , the number of “transferred” NBs at λ = 0 was subtracted from the measured number of transferred NBs and the total number of fired NBs before calculating the transfer probability . We further examined how the transfer probability depends on the strength of inhibitory synapses [41] . For low inhibition levels , the transfer probability ( Fig 8B ) was highly dependent on the inter-cluster connectivity and changed between 0 and 1 ( full transmission ) , indicating that modularity directly controls transmission probability . However , this modulation occurred over a narrow λ range and was insensitive to disinhibition ( eliminating inhibitory synapses analogous to globally applying Bicuculline to the in vitro network ) ( Fig 8B ) . Increasing the decay time constant for inhibitory synapses ( τd ) , or the inhibitory synaptic strength ( A ) twofold , increased both the transition range and the sensitivity to inhibition block ( Fig 8C ) . This suggests that various mechanisms that increase inhibition within the time scale of an NB , such as selective increase in synaptic strength , or neuromodulation of synaptic decay dynamics , can support gating by disinhibition . We next tried to identify other putative mechanisms which can increase disinhibitory effects in modular networks by focusing on the properties of the bridge connecting the two clusters [4] . We set the inter-cluster connection probability between pre-synaptic excitatory and post-synaptic inhibitory neurons to be as high as the excitatory to excitatory probability . This dramatically increased the range over which transmission was modulated ( Fig 8D ) . Furthermore , following disinhibition , the transfer probability dramatically increased ( Fig 8D—red curve ) , opening the gate between the two clusters . Thus , under these settings , our model captured the conditional propagation between sub-populations in a modular network and the gating of this conditional propagation by disinhibition . An alternative to increasing inhibition strength is targeting inhibition , for example , by selectively directing inhibition to excitatory neurons . The existence of such targeting was verified both in the cortex and other brain regions [42 , 43] . Specifically , we examined whether it is possible to control transmission by selectively directing the output of inhibitory neurons , which receive inter-cluster input , to excitatory neurons ( see detailed connectivity schemes in Materials and Methods ) . Indeed , doing so increased the control over transmission to almost the same extent as when increasing the feed-forward inhibition ( Fig 8E ) . Interestingly , the two mechanisms described above were largely insensitive to the elimination of inhibitory connection through the bridge ( Fig 8D and 8E–gray line ) , congruous with the notion of local inhibition . Similar to the in vitro modular networks described above , the propagation of activity between connected sub-populations in the model was characterized by long delays ( calculated as the time lag between the locations of NB peaks in both clusters , see Materials and Methods ) . Delays of several tens of milliseconds decreased with the connectivity between clusters ( Fig 8F—blue curve ) . These delays were independent of the strength of inhibitory drive or whether it was removed altogether , as in the case under inhibitory block , suggesting that they are a consequence of the modular organization ( see Discussion ) . Finally , we examined whether information about the firing rate is transmitted through the bridge between the two sub-populations , as in our in vitro experiments . We analyzed the correlations between the total firing rates in the sending cluster as a function of the firing rate in the receiving cluster ( similar to Fig 3D ) . We found that the correlation increases as a function of modularity ( Fig 8G ) and begins to saturate for λ = 0 . 1 .
In this study , we systematically examined , experimentally and theoretically , the effect of network modularity on activity transmission between neuronal assemblies . A clear hallmark of the modular networks we studied is their capacity to support long delays in the order of 100 milliseconds ( corresponding to a 5 μm/ms velocity between connected clusters ) ( Figs 3E and 6C-middle ) . Since axonal propagation speeds in culture are fast ( >200 μm/ms , [44] ) , the observed delays are likely to be associated with the time it takes the receiving cluster to generate a network response due to multiple synaptic delays ( recruitment time ) ( S3 Fig; for a more elaborate discussion see [39] ) . These delays are shorter on average in large networks of connected clusters , presumably due to the increased number of pathways between any two clusters , but are still much longer than in uniform networks . Why is the capacity to support long delays useful ? Foremost , long delays give rise to time scale separation between the activities of different modules and are a means to dissociate the intra-module from the inter-module processing [21] . Interestingly , our networks ( Fig 3E-right ) showed high variability in the average delays between different networks . This variability may stem from network architecture variability , and implies that modularity has the potential to support variable delays . Indeed , our model shows that delays can be controlled by modifying the coupling between sub-populations ( Fig 8F ) . In a previous report ( performed under the same experimental conditions as here ) we showed that increasing the coupling in modular networks resulted in shorter delays [45] . We also observed variability in delays within specific networks over time ( Fig 3E-left ) , implying that delays may be dynamically regulated to control transmission , for example by short- term plasticity [46] . An additional property of modular circuits is their activation asymmetry . Asymmetry may be important for controlling transmission directionality . In cluster chains , asymmetry was manifested by propagation patterns which were more probable than others ( Fig 4A–4C ) . It was previously reported that coupled networks of similar sizes exhibit inherent asymmetry , and that this asymmetry is associated with the structural asymmetry of the connecting bridge [32] . Accordingly , a small subset of neurons at the bridge controls the propagation between networks . Our results support these previous findings , but suggest that functional asymmetry is also affected by the manner by which the coupled network is embedded within a larger network . When a network of coupled clusters was connected to a larger network , activity mostly propagated from the cluster connecting the large network to its neighbor ( Fig 4E ) . Furthermore , higher asymmetry in the embedding network resulted in higher asymmetry between coupled clusters ( Fig 4E ) . Interestingly , inhibition played a major role in determining asymmetry . Blocking the inhibition drastically affected the propagation direction between connected clusters ( Fig 5E vs . 5F ) , suggesting that transmission directionality can be modulated to a large extent by reducing inhibition . However , further experiments , in which the cluster composition ( e . g . the number of excitatory vs . inhibitory neurons ) is monitored , are required to understand the morphological basis of asymmetry and transmission . Neural networks have to maintain a fine balance between segregated activation ( where activity is restricted to a specific sub-population ) and integrated activation ( where activity spreads to connected sub-populations ) [9 , 47] . In the brain , functional segregation is associated with the structural modularity of the circuit [10] , and the balance between functional segregation and integration explains the high functional complexity in the network [19 , 22 , 48 , 49] . We have shown that a similar fundamental property exists in small modular circuits . Namely , networks support a wide variety of activations from activity epochs which are confined to one sub-population to large-scale activation of the entire network . We observed such features both in one-dimensional cluster chains ( Fig 5A and 5E ) , and in two-dimensional clustered networks ( Figs 6A and 6C-top and 7B and 7F ) , but to a much lesser extent in uniform networks ( Fig 6A , C-bottom ) . We found that both the number and intensity of activated clusters ( Figs 5E , 5G , 7E and 7F ) , as well as the propagation direction ( Figs 5B , 5D , 7B and 7E ) , were highly variable between consecutive NBs . In addition , due to long delays , this spatial activation variability resulted in temporal variability of NB durations . Interestingly , both temporal and spatial variability were previously reported for small-scale circuits in cortical slices . Organotypic slices show avalanche-like patterns typified by a wide distribution ( heavy tail ) of event sizes and durations [50] . The hypothesis that this diversity reflects the transient activation of different cell assemblies is supported by our results . We suggest that the activation diversity in our networks is the outcome of the conditional propagation between sparsely coupled clusters . In small modular networks , activation of one sub-population does not necessarily lead to the activation of the other ( Fig 3B and 3C ) . Our model supports this idea and illustrates how conditional propagation can be controlled by changing the degree or architecture of coupling between sub-populations ( Fig 8C–8E ) . We note that we focus here only on a specific dimension of activation diversity . Uniform networks exhibit rich dynamical behavior along many spatial and temporal degrees of freedom [38 , 39 , 51 , 52] . However , a fundamental property of their activity is network synchrony on a time scale of ~100 ms [39] . Each of our clusters exhibits similar activity profiles to uniform networks [29] , but the weak connections between clusters allows to spatially and temporally decouple their activity on this time scale . We further suggest that the fact that conditional propagation spontaneously emerges in modular networks ( in contrast to uniform networks ) is associated with the networks’ self-regulation . We previously showed that isolated clusters of different sizes , and different connectivity , sustain moderate activity levels [29] . This corresponds to well-documented reports of structural and functional self-regulation of excitability in neurons [41 , 53–55] . Thus , neurons increase or reduce their propensity to fire when activity levels are low or high respectively . However , since in modular networks the fraction of connections between sub-populations is lower than within sub-populations , moderate activity in a cluster may still be below the self-regulated activation limit in the connected cluster ( assuming that all sub-populations employ similar self-regulation mechanisms ) , resulting in the threshold-like behavior we observed ( Fig 3C and 3D ) . Thus , one possible mechanism for transient increase in transmission is by increasing firing rates in the sending cluster , which may be one of the components contributing to transmission under disinhibition . Conversely , the post-synaptic currents to neurons in the receiving cluster at the time of the burst are another factor determining transmission . In principle , such currents can be modulated by neuromodulators or by a third cluster impinging on the receiving cluster . The latter is expected to be more prominent as the number of connections in the network increase . Indeed , while in one-dimensional chains , activity slightly decayed as it propagated between connected pairs , contributing to propagation failure , in two-dimensional clustered networks , a considerable fraction of events still managed to propagate to a large fraction of the network ( S5A Fig ) . This may suggest that transmission in these two-dimensional networks is enhanced by the fact that each cluster is connected to several clusters , thus increasing the probability that their simultaneous activation recruits the receiving cluster . Further studies of single clusters receiving convergent input from two or more controlled clusters are required to explore activity integration in such networks . By design , our modular networks were spatially regular ( Fig 1C ) . Such a design minimized the variability in connections between clusters and variability in cluster sizes ( S1B Fig ) , since self-organization is constrained by the regular pattern . This allowed us to focus on effects of modularity while partially avoiding the influences of other topological properties . For example , it was shown that if clusters freely organize without spatial constraints , cultures drive themselves towards assortative topologies with a “rich-club” core [56] . Such an organization equips modular networks with additional functional features , such as higher resilience to network damage compared to uniform networks . We previously reported [45] that modular networks with strong connectivity between modules do not give rise to long delays and conditional propagation . In addition , uniform networks in cultures are rarely uniform [51 , 52] , and the uniformity probably depends on the level of granularity at which connectivity is examined . For this reason , we focus here on the extreme case of high intra-cluster connectivity with weak inter-cluster connectivity . Only at this level do we see a marked transition in the network’s activity profiles . We showed that disinhibiting networks allows us to control transmission through modular circuits ( Figs 5 and 7 ) . Disinhibiting the network replaced the conditional propagation with reliable transmission ( Figs 5A–5E , 7F and 7G ) . Not only was the diversity in the number of activated clusters removed ( Figs 5E–5H and 7E–7G ) , but also the diversity in propagation directions ( Figs 5E and 5F , 7D and 7E ) . These results were reproduced by our model . Blocking inhibitory synapses increased transmission , effectively “opening the gate” between connected modules ( Fig 8D and 8E ) . Interestingly , this effect was weaker when the inter-cluster connectivity scheme was similar to the intra-cluster scheme ( Fig 8C ) , suggesting that either stronger feed-forward ( Fig 8D ) or targeted ( Fig 8E ) inhibition may be instrumental in controlling propagation . In addition to increased transmission , blocking inhibition in modular networks led to the emergence of an activation focus , which was absent before disinhibition ( Figs 5F and 7B ) . The global disinhibition we induced in our examinations is used to illustrate the capacity of disinhibition to gate the system between different propagation states . In vivo , the inhibition-excitation ratio can be controlled locally , for example by neuro-modulators [57] . Further studies in which the degree of inhibition is manipulated in selective clusters ( for example using Channelrhodopsin and Halorhodopsin ) will determine the degree to which such gating can be controlled . Our global disinhibition was used to establish a proof of principle and is more akin to pathological conditions , as in the case of epilepsy where inhibition control is suspected to fail in large neuronal populations [58] . Indeed , under such conditions , the emergence of a focus ( Figs 5F and 7B ) , and the repeated activation waves ( Fig 5B ) , are clear hallmarks [59] . Interestingly , in addition to inhibition deficiency [60] , lack of sparse functional connectivity between brain modules was also associated with epilepsy [61] . We note that in previous theoretical studies , gating was investigated in the context of balanced networks showing asynchronous irregular patterns [3 , 5] . In contrast , our study targets a different regime of population activity patterns . Primarily , our neurons are not constantly driven as the aforementioned model neurons . In such networks , excitatory-inhibitory balance does not result in persistent irregular patterns , but in synchronized bursting behavior . Nevertheless , excitatory-inhibitory balance does exist in these networks and is vital to the networks’ functionality [41] . Consequently , our model system is relevant for investigating gating during increased activity transients as occurring during synchronized and/or bursting activity . Such transient increase in excitability may have a fundamental role in transferring information between different cell populations [17 , 62 , 63] . To conclude , it is widely accepted that structure and function are closely related in neuronal circuits . However , the contribution of circuit topology to circuit function often remains hidden due to the difficulty in isolating small circuits in intact tissue . By engineering modular circuits in vitro , we explored the functional consequences of modularity and demonstrated that modular topology and disinhibition are instrumental in gating activity , directly demonstrating how structure can shape function in small neuronal circuits .
The entire neo-cortex of ( E18-19 ) Sprague Dawley rat embryos of either sex were removed , chemically digested and mechanically dissociated by trituration , as detailed in a previous publication [29] . Dissociated cells were suspended in a growth medium and plated onto patterned substrates at a density of 700 cells/mm2 . To promote the long-term cell survivability , a “feeder” colony of cells was added to the culture chamber [64] . The surrounding feeder culture did not directly contact the patterned culture . The mitotic inhibitor , FuDr ( 80μM FuDr , Sigma , Cat . No . F0503 and 240μM Uridine , Sigma , Cat . No . U3303 ) was added after four days in culture . Cultures were maintained at 37°C with 5% CO2 and 95% humidity . The growth medium was partially replaced every three to four days . The procedure was done in accordance with the NIH standards for care , and use of laboratory animals and was approved by the Tel Aviv University Animal Care and Use Committee . Overall 69 cluster chains ( from 26 cultures ) , 15 large clustered networks , and eight uniform networks were tested in this study . Extra-cellular recordings were conducted using a low noise pre-amplifier board ( MEA1060-BC amplifier , gain ×1 , 100 with a band-pass filter of 10 Hz to 3 kHz , by Multi Channel Systems , MCS , Reutlingen , Germany ) . Signals were sampled at 10 kHz and stored on a personal computer equipped with a 60 channel , 12-bits data acquisition board ( MC_Card , MCS GmbH ) , and an MC_Rack data acquisition software ( MCS GmbH ) . An additional 200 Hz high pass filter ( 2nd order Butterworth ) was applied to the data stream by the software . Recordings were performed 12 to 28 days in vitro . The patterning method was adapted from a previous publication with slight modifications [27] . Briefly , PDL ( Sigma , Cat . No . p7889 ) islands on top of commercial MEAs ( MCS GmbH ) were prepared with a soft lithography process using polydimethylsiloxane ( PDMS ) stencils . An SU8-2075 ( MicroChem Corp ) mold with approximately 150 μm thickness was casted onto a patterned silicon wafer . The pattern consisted of a rectangular grid ( 6 x 10 ) of circles with diameters ranging between 80 and 200 μm with 500 μm spacing . The PDMS stencil was prepared by spin coating the wafer with PDMS . After detaching the PDMS substrate from the mold , the stencil was placed on commercial MEAs and aligned with the electrode locations . The PDL solution was applied to the PDMS stencil and the PDL was dried on a hot plate at 37°C for half an hour . The PDMS stencil was removed before cell plating . The probability of inter-cluster connections depended on island diameter: Larger islands resulted in networks with a higher degree of connectivity . The network's self-organization lasted up to ten days in culture , after which the patterns became stable . To quantify network level activity , we calculated the activity intensity ( AI ) of each cluster: AI={A , A≥00 , A<0 , A=∑i=1M|V ( i ) |M−NT where V is the voltage waveform , M is the number of samples in each activity intensity bin , and NT is the activity intensity noise threshold . The noise threshold is added to remove the contribution of noise to the AI value and was calculated as follows: The unbiased kurtosis ( measuring Gaussianity ) of the voltage trace is calculated in time bins of 20 ms . The kurtosis of a univariate Gaussian distribution is 3 . Active bins were characterized by super-Gaussian distributions; therefore bins with kurtosis values higher than 3 . 1 were rejected . The waveforms of the rest of the bins were used to estimate the average absolute value of the noise voltage , which is the noise activity intensity threshold , NT . Once NT is obtained , AI is calculated according to the equation for AI ( above ) . We have previously shown that the activity intensity measure can serve as a good estimate for changes in firing rate of superimposed spikes [29] . To detect NBs in single channels we used a previously described method [29] . Briefly , we first calculated AI in bins of 2 ms . Next , we counted the number of active AI bins ( having non-zero values ) in moving windows of length W = 100 ms ( steps of 10 ms ) . W = 100 was chosen because it is long enough to achieve a smooth NBs profile ( but not longer than a typical NB ) . In the model W = 10 was long enough , since the large number of sampled neurons resulted in a smoother profile to begin with . Single sporadic spikes , as well as short threshold crossings , may contaminate NB profiles . As the rate of these events is well below 10Hz , we eliminated them by zeroing data points below a threshold value of T ( T = 10 ) . For the model , this value was chosen as T = 5 since the activity is not contaminated by noise . To ensure that the activity near NB edges was included , a second convolution with the same kernel , followed by thresholding with a value of 1 , was performed . In the resulting time series , NBs are represented by a series of consecutive positive values . Finally , to ensure that short transient decreases did not result in a separation of NBs to two events , NBs occurring less than G milliseconds apart ( end of previous to beginning of next ) were merged into one NB ( G was set to 100 ms for single clusters in the experimental data and 50 ms for the model in which response variability was lower ) . To increase the accuracy of the NB start time , end time and peak time detection , the AI function was extracted during the time windows of the previously calculated NB occurrences . The beginning and end of NBs were taken as the first non-zero value from left and right respectively . The peak of the NB is determined by smoothing the AI ( using a convolution with a Gaussian , σ = 50 ms ) , and extracting the time of the maxima . For identifying NBs globally in the whole network , instead of in single channels , events were counted in all channels instead of only in one , and G was set to 1 s ( in accordance with the large delays in clustered networks ) . In addition , channels with very weak activity during each NB ( active for less than 10 ms ) were not considered . To extract NBs in the computational model , the spike timings of all neurons were counted instead of activity events . The results of NB detection and parameter extraction were verified by manual inspection for all clusters . Detection of correlations between bursts was performed using a hierarchical clustering algorithm [38] . Briefly , AI traces were extracted during a 1000 ms window surrounding the detected NB peaks and smoothed by convoluting with a Gaussian ( σ = 10 ms ) . The time invariant correlation between the ith and jth NBs were calculated as follows: Rij=maxt{⟨Cijn ( t ) ⟩n} , where Cijn is the normalized cross-covariance between the AI trace during the ith and jth NBs of the nth cluster , and t is the time index of the cross-covariance function . To identify groups of similar bursts , Rij is reordered using the dendrogram hierarchical clustering algorithm . The dendrogram is calculated on the Euclidian distance matrix , Dij , between the ith and jth rows in R: Dij2 = ∑k ( Rik − Rjk ) 2 . Simulated networks consisted of two clusters , each with 50 Morris-Lecar neurons ( see neuron model for details ) . Neurons were connected through modified Tsodyks-Markram synapses ( see synapse model for details ) . Connectivity was defined by the matrix aij , where aij = 1/0 corresponds to an existing/non-existing connection between the pre-synaptic terminal of neuron i and the post-synaptic terminal of neuron j . One of every five neurons was randomly selected as inhibitory . Each network was simulated for 300 s using an Euler integrator with a 0 . 1 ms time step . The modularity of the network was determined by the parameter λ as follows . The connectivity probability within cluster 1 ( neurons 1 to 50 ) and cluster 2 ( neurons 51 to 100 ) was initially set to 0 . 25 and 0 . 2 respectively , and the connectivity between clusters was set to 0 . Next , we randomly replaced intra-cluster connections with inter-cluster connections with probability λ . Thus , for λ = 0 the network is composed of two isolated ( disconnected ) clusters , and for λ = 0 . 5 the intra-cluster connectivity is equal to the inter-cluster connectivity , and the initial separation to two clusters can no longer be observed . We simulated three connectivity schemes , which differ by the number and distribution of inhibitory ( I ) and excitatory ( E ) synapses: proportional inhibition , strong feed-forward inhibition , and direct targeting inhibition . We kept the E/I neuron ratio constant , although in the experimental conditions some variability may occur . Such a choice is consistent with the self-regulation of synaptic transmission which compensates changes in the network structure to maintain E/I balance [41 , 65] . In general , the connectivity between two neurons can be one of the following: E→E , E→I , I→E and I→I . These schemes were differentiated by ⟨Nv→w , x→y⟩ which denotes the expected value of the number of synapses from cluster v to w , where x and y stand for the type of the pre and post-synaptic neuron type ( E or I ) . For example , N2→1 , E→I is the number of E → I synapses from cluster 2 to cluster 1 . For the above connectivity schemes , two conditions were used: inhibition block and local inhibition . For inhibition block ( analogues to application of Bicuculline ) , all inhibitory synapses were disabled by setting A = 0 ( see synapse model ) . For local inhibition , only inhibitory inter-cluster connections were disabled ( A = 0 ) . Neurons were modeled as Morris Lecar elements [66]: CmV˙=Iext−gCaMSS ( V−VCa ) −gKW ( V−VK ) −gL ( V−VL ) W˙=ϕ ( WSS−W ) cosh ( V−V32V4 ) MSS ( V ) =0 . 5 ( 1+tanhV−V1V2 ) WSS ( V ) =0 . 5 ( 1+tanhV−V3V4 ) where V is the membrane potential , Iext is the externally applied current , W and M are the fraction of open K+ and Ca+2 channels respectively , and Cm , ϕ , V1 , V2 , V3 , V4 , gk , gCa , gL are constants , adopted with slight modifications from Rinzel and Ermentrout [66] ( see Table 1 for a full list ) . These parameters were selected to simulate a class I neuron which can generate cellular-level [66] and network-level [40] , bursting in accordance with the activity of isolated clusters [29] . We also examined networks of leaky integrate and fire neurons . Although we could qualitatively reproduce our results with these neurons , Morris-Lecar neurons gave a much better fit to the experimental observations ( for discussion see [40] ) . Neurons received both synaptic and noise input: Iext = In + Isyn The noise , In , fed into every neuron , was selected from a Gaussian distribution ( mean , μ=7 . 55[μAcm2] and standard deviation , σ=4[μAcm2] ) ( independently identically distributed for each neuron ) for every simulation step . This choice was driven by the assumption that noise originated from spontaneous synaptic release of neurotransmitter [67] . Such noise could be modeled as an Ornstein–Uhlenbeck process [68] . Considering that for single synapses , the time between spontaneous releases is a Poisson process [69] and that the number of synapses onto a neuron is large , the overall spontaneously evoked noise current can be approximated by a single Gaussian variable ( in accordance with the central limit theorem ) . Synapses were modeled as modified Tsodyks-Markram elements [70]: x˙=z ( −tan ( 1 . 2z−1 . 2 ) ) τrec−uxδ ( t−tAP ) y˙=−yτd+uxδ ( t−tAP ) z˙=yτd−z ( −tan ( 1 . 2z−1 . 2 ) ) τrec where x , y and z are the fractions of synaptic resources in the recovered , active , and inactive states of the synapse . τrec and τd are time constants representing the recovery and decay of active resources respectively ( see Table 2 for a full list ) . tAP is the arrival time of the last action potential to the pre-synaptic terminal . u is the fraction of resources activated upon action potential arrival . In excitatory synapses , u was constant ( u = U0 ) . In inhibitory synapses , u was a dynamic variable enabling synaptic facilitation during bursts: u˙=−uτfacil+U0 ( 1−u ) δ ( t−tAP ) We modified the x and z terms in the original Tsodyks-Markram model by adding a tangent function to prevent tonic endless spiking . Such tonic spiking was observed when the network fired in high rates and originated from the linearity between the recovery rate and the amount of inactive resources . Our modification is in accordance with findings indicating that synapses do not increase their recovery rate following depletion , but rather decrease it after a certain level of depletion [71–73] . The synaptic input to a neuron , Isyn , was calculated by summing over synaptic currents from all connected neurons: Isyn ( t ) =∑iAiyi ( t ) where Ai is the synaptic strength . To reflect the non-uniformity of synaptic strengths , they were selected from a Gaussian distribution ( μ = Anom , σ = Anom/2 ) , where Anom is the nominal value of the synapse [70 , 74] ( see Table 2 for a full list ) . To limit the distribution of strengths , only values between 0 . 8Anom ≤ A ≤ 1 . 2Anom were considered ( values were redrawn from the distribution until a value within limits was drawn ) .
|
The capacity to transmit information between connected parts of a neuronal network is fundamental to its function . The organization of network connections ( the topology of the network ) is therefore expected to play an important role in determining network transmission . Since modular topology characterizes many brain circuits on multiple scales , investigating the role of modularity in activity gating is clearly desirable . By engineering such modular networks in vitro , we were able to perform such an investigation . Under these experimental conditions , we can independently control the degree of modularity , as well as inhibition in the network . We show that a combination of these two properties is highly beneficial from a communication perspective . Namely , it equips connected modules and large modular networks with the capacity to gate and temporally coordinate activity between the different parts of the network .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"neural",
"networks",
"nervous",
"system",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"membrane",
"electrophysiology",
"bioassays",
"and",
"physiological",
"analysis",
"thermodynamics",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"entropy",
"animal",
"cells",
"neural",
"pathways",
"electrophysiological",
"techniques",
"engineers",
"physics",
"people",
"and",
"places",
"professions",
"cellular",
"neuroscience",
"electrode",
"recording",
"neuroanatomy",
"cell",
"biology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"population",
"groupings",
"neurophysiology"
] |
2016
|
Modularity Induced Gating and Delays in Neuronal Networks
|
It is widely believed that innate immune responses to Borrelia burgdorferi ( Bb ) are primarily triggered by the spirochete's outer membrane lipoproteins signaling through cell surface TLR1/2 . We recently challenged this notion by demonstrating that phagocytosis of live Bb by peripheral blood mononuclear cells ( PBMCs ) elicited greater production of proinflammatory cytokines than did equivalent bacterial lysates . Using whole genome microarrays , we show herein that , compared to lysates , live spirochetes elicited a more intense and much broader transcriptional response involving genes associated with diverse cellular processes; among these were IFN-β and a number of interferon-stimulated genes ( ISGs ) , which are not known to result from TLR2 signaling . Using isolated monocytes , we demonstrated that cell activation signals elicited by live Bb result from cell surface interactions and uptake and degradation of organisms within phagosomes . As with PBCMs , live Bb induced markedly greater transcription and secretion of TNF-α , IL-6 , IL-10 and IL-1β in monocytes than did lysates . Secreted IL-18 , which , like IL-1β , also requires cleavage by activated caspase-1 , was generated only in response to live Bb . Pro-inflammatory cytokine production by TLR2-deficient murine macrophages was only moderately diminished in response to live Bb but was drastically impaired against lysates; TLR2 deficiency had no significant effect on uptake and degradation of spirochetes . As with PBMCs , live Bb was a much more potent inducer of IFN-β and ISGs in isolated monocytes than were lysates or a synthetic TLR2 agonist . Collectively , our results indicate that the enhanced innate immune responses of monocytes following phagocytosis of live Bb have both TLR2-dependent and -independent components and that the latter induce transcription of type I IFNs and ISGs .
Lyme disease ( LD ) , the most commonly reported vector-borne illness in the United States , is a tick-borne , multi-system , inflammatory , infectious disorder caused by the spirochetal bacterium Borrelia burgdorferi ( Bb ) [1] . The disease is often heralded in its early stage by erythema migrans ( EM ) , an expanding annular rash which develops following inoculation of spirochetes into the skin at the site of tick feeding and is frequently accompanied by ‘flu like’ symptoms , including myalgias , arthralgias , and fever [1]–[3] . If treated appropriately , the prognosis is excellent [3] , [4]; however , if untreated , hematogenous dissemination of spirochetes may give rise to a wide range of clinical manifestations , most commonly involving the central nervous system , joints and heart [1] . Within days of treatment , the signs and symptoms associated with the disease typically begin to subside , although in some individuals a complete recovery can take several weeks or even months [5] . A minority of treated patients may go on to develop a poorly defined fibromyalgia-like illness , which is not responsive to prolonged antimicrobial therapy [6] , [7] . Understanding the ontogeny of the immune response to the bacterium may provide insights into why some patients remain persistently symptomatic while others recover more rapidly . Until recently , most efforts to understand how Bb initiates innate immune cell activation have focused on the pro-inflammatory attributes of spirochetal lipoproteins [8]–[15] , while less has been done to define the mechanisms underlying immune recognition elicited by live spirochetes . The emphasis on borrelial lipoproteins ( BLPs ) as innate immune agonists emerged from the discovery that spirochetes expresses an abundance of these molecules , many on their outer membrane , and that the borrelial cell envelope lacks the potent gram negative proinflammatory glycolipid , lipopolysaccharide ( LPS ) [16]–[19] . Unlike LPS , which signals through the pattern recognition receptors ( PRRs ) Toll-like receptor ( TLR ) -4 and CD14 [20] , lipoproteins signal through TLR1/2 heterodimers [10] , [13] , [21]–[23] , also in a CD14 dependent manner [12] , [24] , [25] . Intradermal injection of spirochetal lipoprotein analogs ( lipopeptides ) , in both animals [8] and humans [15] , [26] , confirmed in situ that BLPs indeed have the capability to activate macrophages and induce dendritic cell ( DC ) maturation . Collectively , these prior studies led to the viewpoint that innate immune cell activation in LD occurs predominantly through the interaction of spirochetal lipoproteins with CD14 and/or TLR1/2 on the surfaces of macrophages and DCs . The findings that experimentally infected mice deficient in CD14 [25] , [27] , TLR2 [14] , [28] developed even more severe arthritis than their wild type counterparts , and that mice lacking the TLR adapter protein myeloid differentiation factor 88 ( MYD88 ) [29] also developed arthritis , provided the first clear indication that intact spirochetes may employ additional TLRs and/or TLR-independent pathways to induce acute inflammation . Further evidence that spirochetes generate TLR2 independent signals in both humans and mice has been obtained using ex vivo stimulation models . Behera et al [30] demonstrated that blocking of TLR2 on the surface of human chondrocytes eliminated the inflammatory response to purified lipoproteins but not to intact spirochetes . The same group [31] recently confirmed using a mouse model that proinflammatory cytokine production was only partially reduced in TLR2 deficient bone marrow derived macrophages ( BMDMs ) stimulated with live Bb and showed that maximal TLR2 stimulation is dependent on phagocytosis . Using human peripheral blood mononuclear cells ( PBMCs ) we previously provided evidence that phagocytosed live bacteria initiate activation programs in monocytes and DCs that differ both quantitatively and qualitatively from those evoked by cell surface-mediated signals generated by lipoprotein rich bacterial lysates [32] , [33] . The marked increases in IL-1β secretion in response to live bacteria were indicative that phagocytosis of the spirochete generates phagolysosomal signals that result in much greater activation of caspase-1 than could be elicited by cell surface mediated TLR1/2 stimulation by bacterial lysates . In the same experiments heat killed spirochetes were readily phagocytosed and also induced a stronger pro-inflammatory response than bacterial lysates . The greater complexity of the signaling events triggered by internalized spirochetes was further underscored by their ability to induce programmed cell death responses in monocytes . The current study was conducted to further elucidate the multifaceted mechanisms by which live Bb initiates and maintains innate immune responses in human monocytes . We focused on monocytes because of the seminal role these cells have in the recognition and killing of spirochetes during the course of LD [34]–[39] . We began by using microarray methods to compare the global transcriptional responses elicited in PBMCs by live Bb and equivalent amounts of borrelial lysates . The arrays provided unambiguous evidence that live bacteria elicit a proinflammatory response that is not just more intense but also broader , involving genes associated with diverse cellular processes including immune activation , ion transport , protein ubiquitination , and cell damage and repair . The finding that live bacteria induced transcription of interferon β ( IFN-β ) , along with a number of Type I interferon-stimulated genes ( ISGs ) , was of particular relevance given that TLR2-derived signals alone cannot induce type I interferons [40] , [41] . Interestingly , type I interferons have recently been shown to have an important role in the development of arthritis in a murine LD model [42] , [43] . Using isolated human monocytes , we demonstrated that this cell is indeed the primary source of the many unique inflammatory signals engendered by live Bb , and that these signaling pathways , including activation of caspase-1 and induction of IFN-β and associated genes , arise under conditions in which spirochetes manifest no capacity to escape vacuolar confinement and degradation . Collectively our results highlight the ability of LD spirochetes to induce diverse signals which coincide with degradation of spirochetes within phagolysosomes that are both stronger and distinct from those generated by spirochetal lipoproteins and which cannot be entirely attributed to the canonical prototypic TLR1/2 dependent signaling .
We previously showed in a PBMC-Bb stimulation model that phagocytosed live bacteria initiate activation programs in monocytes and dendritic cells ( DCs ) that differ both quantitatively and qualitatively from those evoked by cell surface-mediated signals generated by equivalent amounts of lipoprotein rich bacterial lysates [32] , [33] . We also demonstrated that abrogation of phagocytosis by Cytochalasin D diminished the production of inflammatory cytokines by live bacteria to levels comparable to those induced by lysates [32] . In this study , we used a whole genome microarray method to more completely characterize the transcriptional responses generated by live or lysed spirochetes in the PBMC-Bb stimulation model . Equivalence between input live and lysed Bb used for each individual experiment was confirmed by SDS-PAGE and silver staining ( data not shown ) . PBMCs also were incubated with fluorescent microspheres ( beads ) to evaluate the transcriptional events associated with the cytoskeletal changes that are known to follow the binding and internalization of inert particles [44] . Gene intensity values generated in response to each of the three conditions studied then were compared to those from unstimulated cells . The statistical analysis revealed that 529 genes were differentially regulated in one or more of the three conditions studied ( live and lysed Bb and beads ) . Of the total group of 529 genes , 400 encode molecules with identified biologic functions , and these were selected for further analysis . Based on a ratio determined by directly comparing individual gene transcript intensity values generated from beads , live and lysed vs . unstimulated PBMCs , we considered genes to be either differentially up-regulated ( ratio >1 . 5 ) , differentially down-regulated ( ratio <0 . 7 ) or unchanged . With few exceptions , the transcriptional profile of PBMCs stimulated with beads was very similar to that obtained from unstimulated cells . In contrast , 213 of the 400 genes studied were differentially up-regulated ( Fig . 1A ) and 187 were down-regulated ( Fig . 1B ) in response to either live and/or lysed Bb . Genes within this group of 400 were then classified into four different sub-groups . The first , designated the “core group” , was comprised of 184 genes whose intensity values compared to unstimulated cells were similarly up- or down-regulated in response to either live or lysed spirochetes ( Table S1 ) . The second group included genes classified as being either more intensely or exclusively up-regulated ( Table 1 ) in response to live Bb . The third group contained genes more intensely down-regulated in response to live spirochetes , in some cases exclusively in response to live bacteria ( Table S2 ) . The final group consisted of a small number of genes that were either strongly or solely up-regulated in response to borrelial lysates ( Table S3 ) . The “core group” of genes provided a snapshot of the various innate immune signaling pathways which are triggered from the cell surface in response to intact spirochetes or spirochetal constituents ( lysates ) ( Table S1 ) . Based on prior work , we considered several of these genes to be largely representative of downstream TLR2-mediated signals . Within this core group there also were several chemokine associated genes , including IL-8 and CCL2 . IL-8 has been shown to be secreted in response to TLR1/2 mediated signals induced by either spirochetal lipoproteins [45] , [46] or Bb lysates [47] . Moreover , synthesis of this chemokine in response to borrelial lysates can be partially abrogated by TLR2 blockade [47] . The TLR-induced neutrophil chemoattractant CCL2 ( also known as MCP-1 ) is over expressed in LD erythema migrans ( EM ) dermal infiltrates [48] and in joint tissues from Lyme arthritis susceptible mice [49] . The gene for the suppressor of cytokine signaling protein 3 ( SOCS3 ) , a negative regulator of cytokines that signal through the Janus kinase/signal transducer and activator of transcription ( JAK/STAT ) pathway [50] , also was induced in response to both stimuli . Of interest , decreased SOCS activity in CD14-deficient murine macrophages has been associated with an exaggerated TLR mediated pro-inflammatory cytokine response to phagocytosed Bb ( personal communication from Dr . Tim Sellati ) . A surprisingly large number of genes associated with diverse metabolic functions were also contained in this core group . This finding indicates that immune cells have the capacity to down-regulate various genes associated with metabolic functions which are not critically needed during inflammatory responses . Some genes which encode cell surface receptors associated with immune signals were also down-regulated . This was the case for CCR2 , which is linked to the cell surface chemokine receptor for MCP-1 [51] , and for IFNGR1 , which encodes the IFN-γ receptor-1 protein [52] . Of note , mouse macrophages TLR2 stimulation with a synthetic ligand caused a similar decrease in IFNGR-1 transcription [52] Conversely , NALP12 ( Monarch1/PYPAF7 ) which is a negative regulator of TLR-induced inflammatory responses [53] was down-regulated by both stimuli . As already noted above , a large number of genes were either exclusively or more intensely up-regulated by live as opposed to lysed Bb . Within this group we found several genes that encode for monocyte-derived chemokines ( CCl3L1 , CCL3 , CCL3L ) [54] , as well as two monocyte/macrophage-associated growth factors ( GMCSF and GCSF ) [55] . This group also encompassed a number of genes which are known to be directly or indirectly associated with phagocytosis and/or assembly of phagolysosomes . PI4K2B , a cytosolic phosphoinositol kinase which regulates vesicular trafficking during phagocytosis [56] and LAMP3 ( lysosome associated membrane protein 3 ) , a late endosomal maturation marker [57] , were both strongly up-regulated . TNF-α and IL-6 , which encode for two cytokines previously shown to be secreted in large quantities in response to live Bb in the PBMC stimulation model [32] , were also included in this group . In that prior study we also showed that phagocytosed Bb generated signals that activated natural killer ( NK ) -cells inducing them to produce INF-γ . Herein , the transcript for INF-γ was more than 40-fold higher in PBMCs exposed to live spirochetes . Correspondingly , the genes for CD69 [58] and SLAMF7 ( also known as CS1 ) [59] , which are both associated with NK cell activation , also were intensely up-regulated by live bacteria . As a likely consequence from the paracrine effects of INF-γ , STAT1 was robustly up-regulated by live spirochetes . In agreement with this assumption , the gene for NMI ( N-myc STAT1 interactor ) which is directly involved in STAT-dependent transcription of IFN-γ [60] , was exclusively regulated by live Bb . The transcript for FXYD6 , which codes for a protein that regulates Na , K-ATPase channels by altering their affinity for Na+ and K+ ions [61] , was up-regulated only by live spirochetes . The latter is particularly noteworthy given that phagocytosed Bb markedly enhanced secretion of active IL-1β by monocytes [32] , a process which first requires TLR stimulation trailed by a series of complex signaling events that lead to assembly of the inflammasome , activation of caspase-1 and cytosolic cleavage of pro-IL-1β [62] . Changes in cytosolic ionic composition following phagocytosis , particularly loss of intracellular K+ , could thus provide the necessary signals required to activate the NALP3 inflammasome [63] , [64] . Live spirochetes also more intensely or exclusively regulated several genes associated with cell damage and repair than bacterial lysates . This finding is in accord with our prior demonstration that phagocytosed Bb , but not bacterial lysates , has the capacity to initiate programmed cell death responses in human monocytes [33] . Gene transcripts within this group included PARP9 ( Poly ADP-ribose polymerase9 ) ; PARP10 , PARP12 , PARP14 and Galectin9 ( LGALS9 ) . The catalytic activity of PARPs has been shown to be stimulated by DNA strand breaks which occur during programmed cell death . Within this cluster , PARP9 and PARP10 were exclusively up-regulated by live Bb . These two PARPs have not been previously associated with apoptosis [65] , suggesting that phagocytosis of Bb may initiate novel mechanism of cell damage and repair . Galectin9 is a β-galactosidase binding lectin which is present in activated macrophages and DCs [66] , and has the capacity to incite apoptosis via the calcium-calpain-caspase-1 dependent pathway [67] . Canonical TLR1/2 signals are not known to promote transcription of type I interferons [40] , [41] . It was , therefore , particularly noteworthy that live Bb induced transcription of IFN-β and several type I interferon-associated genes in human PBMCs ( Table 1 ) . Within this cluster there were multiple interferon-induced protein transcripts ( Mx1 , IFIT1 , IFIT2 , IFIT3 , IFI16 and RSD2 ) , as well as several interferon-regulated genes known to code for molecules involved with protein ubiquitination or ubiquitin-like functions ( ISG15 , HERC5 , UBE2L6 and USP18 ) . The transcript for ISG15 , which featured prominently within this cluster , codes for a ubiquitin-like molecule that is conjugated to intracellular target proteins after type I interferon stimulation [68] . HERC5 is an IFN-β-inducible E3 protein ligase [69] which localizes to the cytoplasm and perinuclear region of cells and is required for ISGylation of ISG15 [68] . UBE2L6 is a conjugating enzyme also responsible for protein ISGylation , while USP18 is a protease that specifically removes ISG15 [70] . The gene for ISGF3G ( also known as IRF9 ) , a translational regulator known to associate with phosphorylated STAT1/2 heterodymers to form a complex termed ISGF3 transcription factor , also was up-regulated by live Bb . ISGF3 enters the cell nucleus and binds to the IFN-stimulated response element ( ISRE ) to activate the transcription of interferon stimulated genes [71] . IRF7 , which was up-regulated by live Bb but not by lysates , is expressed constitutively in monocytes and DCs but can be induced following single-stranded RNA mediated activation of endosomal TLR7 or TLR8 [72] . IRF7 also can be up-regulated by IFN-β through a cell surface initiated paracrine loop that activates STAT1 [73] . Several genes associated with cell signaling , ion and metal transport systems , and cytoskeleton architecture were more intensely or exclusively down-regulated by live Bb ( Table S2 ) . The gene for TLR6 , which together with TLR1 recognizes diacylated lipoproteins [74] , was intensely down-regulated by both stimuli , but much more so by the live spirochetes . Since Bb does not have diacylated lipoproteins on its outer membrane [16] , this finding suggests that inflammatory cells are capable of fine-tuning transcriptional responses depending on the biochemical configuration of the lipoproteins encountered . The transcript for CD91 ( LRP1 ) , a transmembrane receptor which is known to bind the anti-apoptotic molecule α-2-microglobulin and the heat shock protein gp96 [75] , was also in this group . This suggests that in response to live Bb innate immune effector cells shift inflammatory responses away from those that are associated with exogenous damage-associated molecular patterns ( DAMPs ) . A small group of genes were more intensely or exclusively up-regulated by bacterial lysates ( Table S3 ) . Of particular interest we found three genes that encode for matrix metallopeptidases ( MMP1 , MMP10 and MMP19 ) . MMP-9 has been previously shown to be induced in both human and murine monocytic cells in a TLR2 dependent manner [76] , [77] . MMP10 was also previously found to be up-regulated in mouse cells infected with live Bb [78] . Several genes associated with structural molecules , including two which are involved in keratin formation ( KRT2A and KRTAP17 ) , were exclusively up-regulated by the lysates . Similar structural genes have been shown to be up-regulated in joint tissue obtained from a murine Bb induced arthritis model [79] . We previously demonstrated by quantitative real time reverse transcriptase PCR ( qRT-PCR ) that the transcripts for TNF-α , IL-6 and IFN-γ were more intensely up-regulated by live Bb than bacterial lysates [33] . As shown in Table 2 , herein we also confirmed by qRT-PCR that live bacteria induced IRF-7 , IFN-β , STAT-1 , and LGALS9 . Although the transcript for galectin-9 was confirmed to be up-regulated by live spirochetes , exclusive down-regulation of this transcript by the lysates as shown in the array could not be corroborated by qRT-PCR . Excluding LGALS9 , the live to lysed transcriptional values determined for each individual gene correlated well between the two methods used ( R2 = 0 . 99 , 95% CI: 0 . 93–0 . 99 , p<0 . 001 ) . IL-1β which was intensely up-regulated by live Bb stimulated PBMCs , was not significantly regulated in the PBMC array analysis . On the other hand , as will be discussed below , several type I interferon genes differentially up- or down-regulated in the array were identically regulated in isolated monocytes stimulated under comparable conditions ( Table S4 ) . Biologic validation for the transcriptional responses denoted in the PBMC array also could be ascertained from prior experiments where we utilized the Bb-PBMC stimulation model [32] , [33] . For instance the revelation that Bb induced programmed cell death in monocytes is in accord with the finding herein that several genes associated with apoptosis were up-regulated by live spirochetes . The sizable increase in the transcript for CD69 shown in the array is also consistent with the prior demonstration that NK-cells express this cell surface activation marker when exposed to live Bb [32] . The transcriptional responses in the PBMC whole genome array , in conjunction with results from our prior ex vivo stimulation studies [32] , [33] , provided substantial evidence that phagocytosis of live Bb triggers inflammatory responses in PBMCs that differ quantitatively and qualitatively from those generated by bacterial lysates at the cell surface . In this study we used the isolated monocytes to verify that this cell was indeed a major source of the transcriptional responses generated by live Bb and that the enhanced responses elicited by viable spirochetes were not dependent on signals or cytokines derived from other mononuclear cells in the PBMC mixture . We first had to verify that uptake of the spirochetes by highly purified human monocytes was equivalent to the uptake previously shown in the PBMC-Bb stimulation model . Not surprisingly , a large percentage of the monocytes ( Mean 63%: SE+/−9 . 6% ) cultured with Bb ( MOI 100∶1 ) contained fully or partially degraded fluorescent bacteria ( see microscopy section below ) . Additionally , a similar percentage of CD14+ monocytes were also GFP+ by flow cytometry ( data not shown ) . We then confirmed that live Bb was significantly more potent than borrelial lysates for inducing transcription and secretion of IL-1β , TNF-α and IL-6 ( Fig . 2 ) . Live spirochetes also induced higher levels of IL-10 ( data not shown ) ; indicating that the enhanced response associated with phagocytosed spirochetes also includes the synthesis of anti-inflammatory cytokines which others have shown to be important for control of LD [42] , [80] , [81] . The marked increases in the amount of secreted IL-1β suggest that phagocytosed live spirochetes induce signals resulting in the activation of caspase-1 , which is a known mechanism for cleavage of pro-IL-1β [82] . Additional evidence for caspase-1 activation was obtained by the demonstration that live Bb also induced monocytes to secrete IL-18 , which though constitutively expressed also requires proteolytic processing by activated caspase-1 [83] . Of particular importance , neither bacterial lysates nor the TLR-4 ligand LPS were capable of inducing monocytes to secrete IL-18 ( Fig . 3 ) . Alternatively , IL-1β could be secreted in response to other caspases which are linked to the initiation of apoptosis . One of the most significant findings from the PBMC array analysis , and subsequently substantiated by qRT-PCR ( Table 2 ) , was the discovery that live Bb induced transcription of type I interferons . As shown in Fig . 4 , live Bb was also capable of inducing a distinct increase in transcription of IFN-β in isolated human monocytes , while a similar response was not observed with borrelial lysates or a high concentration of a synthetic TLR2 ligand ( MMP 50 µg/ml ) . The finding that LPS induced transcription of IFN-β is in line with prior observations demonstrating that TLR4 activation can generate type I interferons through the MyD88-independent TRIF-dependent pathway [84] . To better characterize the breadth of the type I interferon responses generated by live or lysed spirochetes in monocytes , we then utilized a Type I interferon RT2 profiler array . This method allows simultaneous , quantitative measurement of 84 Type I interferon associated gene transcripts . Although some overlap did occur in the type I interferon transcriptional responses generated by live and lysed Bb , without exception the response was of greater intensity in monocytes stimulated by live bacteria ( Fig . 5A and 5B and Table S4 ) . Live Bb not only exclusively induced transcription of IFN-β , but also transcription of the gene that encodes for IFN-κ , whose functional profile resembles that of IFN-β [85] . The gene for the interferon inducible CXCL10 ( IP-10 ) , which encodes the ligand for CXCR3 , also was significantly up-regulated by live Bb . Increases in CXCL10 mRNA have been shown in dermal lesions [86] and joint fluid [87] of LD patients . Several other type I interferon inducible genes including ISG15 , IFIT1 , IFIT2 and IFIT3 also were strongly up-regulated by live Bb in monocytes , as in the PBMC system . IL-6 was induced by both live and lysed Bb , which is not surprising since this cytokine is known to be secreted in response to stimuli other than those associated with type I interferons; including TLR2 ligands as well as borrelial lysates [33] . The unique transcriptional responses and cytokine outputs demonstrated here highlight two distinct consequences resulting from the interaction of Bb with human phagocytic cells: ( 1 ) a markedly intensified TLR mediated pro-and anti-inflammatory cytokine output , ( 2 ) activation of signaling pathways generally associated with intracellular bacteria ( i . e . Listeria monocytogenes and Franciscella tularensis [88] , [89] ) and some extracellular pathogens ( i . e . group B streptococcus [90] ) , which can escape the confines of the phagosome to trigger cytosolic inflammatory responses . Although phagocytic cells have been shown to internalize and degrade Bb [32] , [33] , [35] , [39] , [91] , consideration of the signaling issues raised above prompted us to visually re-examine the fate of the spirochete when it comes into contact with the monocyte . Epifluorescent microscopy revealed that individual monocytes contained several fluorescent vacuoles with either bacterial coils and/or partially or fully degraded spirochetes ( Fig . 6A ) . Of interest , and in concert with our prior demonstration that live Bb induces programmed cell death in monocytes [33] , cells also exhibited various stages of nuclear fragmentation ( Fig . 6A ) . This finding , which is traditionally associated with apoptosis , correlated with a dose dependent decrease seen in monocyte counts in response to live Bb ( data not shown ) . Because it was not always possible to determine if intact spirochetes were located intra- or extracellularly using epifluorescent images , we then used confocal microscopy which is better suited for such purpose ( Fig . 6B–F ) . The representative horizontal optical slices shown in the figure reveal two intact spirochetes that are in close association with the monocyte and several degraded bacteria contained within intracellular vacuoles . Colocalization of GFP fluorescent bodies with Lyso-Tracker dye ( red ) provided evidence that some of the digested bacteria were inside phagolysosomes . To determine the precise location of intact spirochetes in relation to the intracellular vacuoles , optical slices were then assembled into stacks and cut perpendicularly ( y , z axes ) and transversally ( x , z axes ) to the imaged planes generating an orthogonal view of the spirochetes . The resulting side and top views of several hundred reconstructed images allowed us to conclude that intact spirochetes , when present , were not within the cell cytosol . The finding that live Bb generates greater transcription and secretion of pro-inflammatory cytokines than similar amounts of bacterial lysates can have two possible explanations . One is that TLR2 signaling proceeds more efficiently or intensely when spirochetes are phagocytosed . The other is that induction of the cytokine responses resulting from internalization of Bb is not exclusively TLR2-dependent . Both of these possibilities are supported by prior work demonstrating that upon phagocytosis of microbial pathogens or bacterial products , TLRs that are on the cell surface can also be recruited to the phagosome and thus become available for signaling [92] , [93] . The most straightforward approach to distinguish between these two possibilities was to examine the cytokine responses to Bb in TLR2−/− murine macrophages . Because the source of the mouse macrophages ( bone marrow vs peritoneal derived ) has been previously shown to be linked to differing uptake as well as cytokine outputs , in this study we elected to use each cell lines under separate experiments . Consistent with prior observations [31] we demonstrated by using flow cytometry and confocal microscopy that uptake and degradation of spirochetes was not significantly affected despite the absence of TLR2 ( Fig . 7A and 7B ) . WT and TLR2−/− macrophages were then stimulated with live or lysed spirochetes ( MOI 10∶1 ) to appraise the output of selected cytokines . Compared to their WT counterparts , macrophages harvested from TLR2-deficient mice and stimulated with live Bb secreted only moderately diminished amounts of TNF-α ( ∼25% less ) ( Fig . 7C ) . Decrease responses were slightly more pronounced for IL-6 ( 1611 pg/ml vs . 604 pg/ml ) and IL-10 ( 248 pg/ml vs . 58 pg/ml ) . In contrast , at similar MOIs the response to lysates was virtually eliminated in TLR2−/− cells ( Fig . 7D ) . While bacterial sonicates have numerous structural components available for cell signaling , the severely impaired cytokine output in TLR2−/− macrophages indicates that the response to the lysates is principally due to spirochetal lipoproteins . Although transcription of IL-1β in TLR2−/− macrophages stimulated with live Bb was not as robust as compared to the WT cells , the absence of TLR2 once again did not eliminate the response . Whereas induction of IL-1β in TLR2−/− cells was absent upon stimulation with bacterial lysates . Unlike human monocytes which were capable of secreting large quantities of IL-1β in response to live bacteria , neither WT nor TLR2 deficient murine macrophages were able to secrete detectable amounts of this cytokine ( data not shown ) . On the other hand , phagocytosed live Bb ( MOI 10∶1 ) was able to similarly induce transcription of IFN-β in both WT and TLR2 deficient bone marrow derived macrophages ( Fig . 8 ) . TLR5 can be expressed in endosomal structures [94] , [95] and can be activated by bacterial flagellin to induce synthesis of pro-inflammatory cytokines [96] . Although not exposed on B burgdorferi's outer membrane , flagellin is a major constituent of the spirochete's periplasmic flagellar structures [97] and thus can become accessible for endosomal TLR5 signaling following degradation of the spirochete within phagolysosomal vacuoles . In concert with this premise , two previous animal studies provided evidence that TLR5 may be partially responsible in generating pro-inflammatory responses to phagocytosed Bb [31] , [95] . Thus to examine the contribution of TLR5 ligation with TLR2 mediated signals , herein we measured human monocyte cytokine output in response to a Bb strain known to be deficient in flagellin [98] . Despite the elongated structure , the mutant spirochetes were readily phagocytosed and degraded within phagolysosomes ( data not shown ) . The flagellin deficient spirochetes were also capable of inducing human monocytes to secrete very similar levels of IL1-β , TNF-α , IL-6 and IL-10 ( Fig . 9 ) , than elicited by wild type Bb ( Fig . 2 ) . This finding indicates that when TLR2 is available for signaling , TLR5 does not appear a play a significant role in production of cytokines in response to phagocytosed live Bb .
In the course of natural infection with the extracellular pathogen B . burgdorferi phagocytic cells are considered to be the first-line of host defense against the bacterium [2] , [25] , [32] , [33] , [99] . Immune cell activation by the spirochete has generally been ascribed to outer membrane lipoprotein-TLR1/2 mediated inflammatory responses [8]–[15] . Evidence ascertained herein from the PBMC array , and subsequently corroborated using similarly stimulated monocytes , corroborated the important contribution of Bb-cell surface TLR1/2 mediated activation in response to the spirochete . Our study results also make obvious that a far more intense and diversified innate immune response coincides transcriptionally and microscopically with phagocytosis and degradation of live spirochetes and maturation of the phagosome . Most prominently , the innate immune signals generated by phagocytosed live Bb led to an enhanced TLR-mediated pro- and anti-inflammatory cytokine output , secretion of active IL-1β and IL-18 , as well as induction of type I interferons . TLRs continuously sample the extracellular environment and inform the cell to react to PRRs by facilitating cellular responses via inflammatory pathways which culminate in cytokine production and cell activation [11] , [41] . Human macrophages , which originate as monocytes in the peripheral blood , express a substantial complement of both cell surface as well as endosomal TLRs [100] . These cells thus have the capacity to sense B . burgdorferi's extensive lattice of outer membrane lipoproteins [23] , [101] . Although the stimulation experiments were not done under conditions designed to prevent uptake of live spirochetes or bacterial components , the transcriptional responses elicited by both live and lysed Bb in PBMCs , for the most part were representative of cell surface TLR1/2 mediated activation . These responses were perhaps best exemplified by the differential regulation of IL-8 as well as several other chemokines . IL-8 is known to be secreted in response to purified spirochetal lipoproteins [45] , [46] as well as Bb lysates [47] . Surface signals were also capable of up-regulating several genes associated with molecules that regulate TLR responses , including PI3K [102] , [103] and SOCS3 [50] . Interestingly , PI3K is also associated with non-opsonic phagocytosis [102] and thus may play an important role facilitating spirochetal binding to the bacterium's putative cell surface phagocytic receptor . Overall , these core responses suggest that TLR-cell surface activation , probably in concert with the engagement of the spirochetes putative phagocytic receptor , set the stage for the more intense TLR-dependent and -independent responses generated by phagocytosed spirochetes . The enhanced TLR-mediated cytokine production could be the result of several nonexclusive mechanisms broadly divided into three categories; ( 1 ) a more efficient activation of recruited cell surface and endosomal TLR1/2 receptors by spirochetal lipoproteins , ( 2 ) engagement of additional endosomal TLRs by internalized and degraded spirochetes , and ( 3 ) cooperation between multiple TLR receptors from within the phagosomal vacuole . Although TLR-PAMP interactions were originally studied as cell surface phenomena; it is now well documented that surface TLRs , including TLR2 and TLR5 , can be recruited to endosomal membranes where they become available for signaling [79] , [92] , [93] , [95] , [104] . The visualization of bacterial coils contained within phagosomal vacuoles provided a snapshot for the intimate physical interactions that very likely take place between the spirochete and vacuolar structures ( Fig . 6A ) . The very close proximity between spirochetal lipoproteins and recruited TLRs can be envisioned as a mechanism that facilitates activation of TLR receptors . Following degradation of the bacterium , liberated lipoproteins then would also be available to more efficiently engage of their cognate TLR receptors . Because TLR2 did not appear to be necessary for phagocytosis , or critically required for cytokine secretion in response to live spirochetes , we propose that other TLRs are involved in generating these responses . Two previous studies , one using stimulated murine macrophages and RAW cells , and the other rhesus microglia , demonstrated that TLR5 signals do contribute to cytokine production in response to phagocytosed Bb [31] , [95] . Our demonstration that the flaA mutant Bb induced similar levels of TNF-α , IL-1β , IL-6 and IL-10 in human monocytes , compared to its wild type counterpart , suggests that as long as TLR2 is available for signaling , TLR5 is not necessary for enhanced cytokine production . Consistent with this theory , silencing of TLR5 in a recent study had no effect on the production of TNF-α , IL-8 , or IL-6 by a monocytic cell line stimulated with live Bb [105] . TLR7 , TLR8 and TLR9 , all of which are known to be expressed in endosomal membranes [106] , may also play an important role in generating the enhanced cytokine responses to internalized Bb . In the end , spirochetes almost certainly engage multiple TLRs concurrently from within the phagosomal vacuole , a conjecture that has been previously demonstrated to occur in response to other bacterial infections [107] . Particularly important for the development of the concept of phagosomal signaling was the markedly enhanced secretion of IL-1β and IL-18 in response to live Bb . Unlike other pro-inflammatory cytokines ( i . e . TNF-α ) , which are linearly induced by TLR activation , the production of biologically active IL-1β requires the integration of NF-κB-mediated transcription of pro-IL-1β followed by activated caspase-1 cleavage of the inactive cytokine [62] . Unlike pro-IL-1β , pro-IL-18 is constitutively expressed in resting monocytes and macrophages [83]; however , like pro-IL-1β it also requires processing into its active form by activated caspase-1 . Caspase-1 is activated within a multiprotein complex called the inflammasome [89] , [108] in response to a diverse stimuli including intracellular bacteria [89] , uric acid crystals [108] , toxins , and changes in the ionic composition of the cell [63] , [64] . Stimulation of cell surface purinergic receptors by exogenous ATP , following human monocyte activation , is a known mechanism by which bacterial pathogens can lead to cleavage of caspase-1 [64] . Released ATP engages the cell's purinergic ion channel receptors ( P2X7 ) inciting release of intracellular K+ which in turn generates signals that lead to assembly of the inflammasome and activation of caspase-1 . Consistent with this theory , the transcript for FXYD6 , which encodes for a protein that regulates Na+ , K+-ATPase channels by altering their affinity for both ions [109] , was exclusively up-regulated in PBMCs stimulated by live spirochetes ( see Table 1 ) . In similarly stimulated mouse macrophages Bb also induced far greater transcription of IL-1β than bacterial lysates; however unlike human monocytes , murine macrophages were unable to secrete the active cytokine . The disparity in IL-1β secretion between mouse and human cells is due to the inability of murine macrophages to produce endogenous ATP following their activation [110] . These differences not only highlight the importance of studying inflammatory responses in human cells , but also provide indirect evidence for the potential role of P2X7 mediated activation of the inflammasome by phagocytosed Bb . Alternatively , caspase-1 could be activated in response to bacterial flagellin leaking from the phagosome into the cell cytosol . Two prior studies demonstrated that flagellin deficient Salmonella typhimurium [111] and Legionella pneumophila [112] failed to activate the inflammasome , thus providing clear evidence that flagellin needs to gain access to the cytosol in order to directly activate assembly of the inflammasome . This mechanism is unlikely to be responsible for activation of caspase-1 in the case of Bb , first and foremost because the spirochete does not have the required cellular machinery to secrete noxious molecules into its surrounding environment . Furthermore , the high concentrations of secreted IL-1β in response to flagellin deficient Bb provides further proof that this molecule is unlikely to be directly responsible for activation of caspase-1 . In other models activation of caspase-1 was achieved by intracellular bacteria that have the ability to escape unscathed from the phagosome into the cytosolic compartment to directly engage cytosolic receptors and activate the inflammasome [113] , [114] . Although the translocation of small amounts of spirochetal components cannot be ruled out in the current study , the evidence presented here demonstrates that intact Bb remains enclosed within phagocytic vacuoles . Interestingly , secretion of IL-1β was greatly impaired in human monocytes infected with a Francisella strain when its natural ability to escape the phagosome was blocked experimentally [115] . Why Bb generates signals that lead to activation of caspase-1 in human monocytes from within the phagosome , while other pathogens do not , requires further analysis . A principal and novel finding in our study was that type I interferons were differentially regulated in both PBMCs and isolated human monocytes stimulated with live Bb . Type I interferon associated genes were previously shown to be strongly up-regulated in joint tissues of Bb infected mice [43] , [79] . The same group provided experimental evidence that type I interferons probably play a very important role in the development of arthritis [43] . These responses are of particular relevance given that lipoprotein mediated TLR2-derived signals do not induce type I interferons [40] , [41] . Several other TLRs , including TLR7 , TLR8 ad TLR9 , are able to launch distinct signaling pathways from within phagosomal vacuoles that differentially regulate type I interferons [116] . Intracellular bacterial deoxycytidylate-phosphate-deoxyguanylate ( CpG ) -DNA can induce type I interferon transcription through TLR9 [117]; however TLR9 is only expressed at very low levels in human monocytes [106] . Both TLR7 and TLR8 detect ssRNA [100] and thus could also explain the type I interferon responses to Bb upon release of bacterial RNA from degraded spirochetes in the phagosomal vacuole . Human TLR7 is predominantly expressed in lung , placenta and spleen , whereas TLR8 is more abundant in peripheral blood leukocytes , including monocytes [100] . Endosomal TLR8 activates IRF7 via MyD88-dependent−pathways involving IRAK1/4 and TRAF6 [72] . Although IRF7 was strongly up-regulated by live Bb in the PBMC array , we were unable to confirm a similar up-regulation of this interferon regulator in human monocytes ( data not shown ) . The latter result could be an indication that in the PBMC model dendritic cells are the principal source of IRF7 . Of note , transcription of type I interferon associated genes in response to Bb was recently found to be MYD88 independent in stimulated murine derived bone marrow derived macrophages [43] . Whether or not transcription of type I interferons in response to Bb is MYD88 dependent in human cells has not yet been fully characterized . Type I IFNs had generally been associated with antiviral immune responses . More recently , an increasing body of evidence points out that induction of these cytokines also occurs in response to infection with both intracellular [88] , [118] , [119] and extracellular bacteria [90] , [120] . IFN-inducing bacterial ligands are primarily detected , with few exceptions , following entry of the bacterium into the cell cytosol . For the most part , the specific cytosolic receptors activated to generate the type I interferons are not known . The intracellular bacterium L . monocytogenes induces type I interferons through an MYD88-independent pathway [88] , [116] , and to generate this response it requires Listeriolysin O ( LLO ) mediated escape from the phagosome into the cell cytosol [116] , [121] . Neither the ligand presented by cytoplasmic LM nor the receptor associated with the type I interferon response has been fully characterized . Streptococcus agalactiae ( GBS ) , an extracellular pathogen associated with severe perinatal infections , also induces type I interferons in human monocytes [90] . However , unlike Bb , GBS can generate toxins that affect the integrity of the phagosome allowing bacterial components to escape into the cytosol to engage cytosolic receptors . Released GBS DNA can activate the serine-threonine kinase TBK1 causing phosphorylation of IRF3 and induction of IFN-β . In the case of Streptococcus pyogenes ( GAS ) , induction of type I interferons was shown to be MYD88-independent [120]; and like Bb , GAS did not require escape into the cytosol to generate this response . Whether or not Bb induced type I interferon responses are initiated from within the phagosome , or by signals generated by released bacterial products into the cell cytosol is not known at this time . It is also not known if Bb initiated programmed cell death responses have a role in generating type I interferons through cross priming of cytosolic receptors . Collectively our results highlight the ability of phagocytosed Bb to induce diverse and more intense innate immune signals which are mechanistically distinct from those generated when spirochetal lipoproteins engage cell surface PRRs . They also demonstrate that human monocytes are a major source of the transcriptional responses generated by live Bb in a mixed cell system ( PBMCs ) , and that these responses are not dependent on inflammatory signals or cytokines derived from other immune cells . It is our contention that the phagosomal signals generated in response to live Bb allow the host to control the spirochete through a number of non-exclusive pathways , that are both TLR2 dependent and independent , and include a type I interferon response . Whether or not the type I IFN response is favorable to the human host , or detrimental as in the mouse arthritis model , remains unknown and deserves additional study .
Eligible participants were healthy volunteers of either sex , between 18 and 60 years of age , and without a clinical or prior history of reactive laboratory tests for LD . After obtaining written informed consent , blood was collected by the University of Connecticut Health Center's ( UCHC ) , General Clinical Research Center ( GCRC ) personnel using standard venipuncture techniques . Volunteers were confirmed to be sero-negative for LD by standard serological tests performed by the UCHC clinical laboratory . Individuals were considered ineligible if they had underlying chronic diseases , were acutely ill , and/or were taking anti-inflammatory medications or any form of immunosuppressive agents . All procedures involving human subjects were approved by the Institutional Review Board at UCHC . Low-passage Bb 297 transformed with a shuttle vector harboring the gene for green fluorescent protein ( GFP ) constitutively expressed from flaB promoter [122] , was propagated in commercially available Barbour-Stoenner-Kelly ( BSK ) complete medium containing 6% rabbit serum ( Sigma-Aldrich , St . Louis , MO ) which is certified to be endotoxin free . Spirochetes grown at 33°C were temperature-shifted to 37°C prior to use; organisms were harvested at mid- to late-log phase ( 4–8×107/ml ) by centrifugation at 8 , 000×g , washed twice in CMRL ( Invitrogen , Carlsbad , CA ) , and resuspended in RPMI medium ( Invitrogen ) . A high passage strain B31 flaB mutant ( 1571 ) , which completely lacks the filamentous portion of the flagellar apparatus ( 44 ) , was grown and harvested in similar fashion . Spirochetal lysates were prepared by sonicating live organisms for 40 seconds using four separate 10-s bursts with a 550-Sonic Dismembrator ( Fisher Scientific , Pittsburgh , PA ) . The equivalence of live and lysed spirochetes was confirmed for each experiment by SDS-PAGE and silver staining . Freshly isolated PBMCs were obtained by Ficoll density gradient centrifugation from three healthy volunteers as previously described [33] . PBMCs were then incubated 4 hours at 37°C/5% CO2 with either live or lysed Bb 297 at an MOI of 10 , with 1 µm Fluoresbrite carboxylated polystyrene microspheres ( Polysciences , Inc . , Warrington , PA ) at a bead-to-cell ratio of 10 , or without a stimulant . At the conclusion of the incubation period , cells were harvested and RNA was extracted using TRIzol according to the instructions of the manufacturer ( Invitrogen ) . RNA samples were then submitted to the Translational Genomics Core facility at UCHC for gene expression profiling using the Illumina Sentrix Human-6 Expression BeadChip microarray system ( Illumina Inc . , San Diego , CA ) . A total of 500 ng of RNA per stimulation condition was converted into complimentary DNA ( cDNA ) and labeled according to the manufacturer's instructions and then assayed in triplicate for each stimulation condition . For statistical analysis , raw gene intensity values generated for each transcript were background-corrected and normalized using the Qspline option in the Beadstudio software ( Illumina Inc . ) . Statistical analysis was conducted using the R/MAANOVA open source software ( Version 1 . 4 ) as part of the Bioconductor and R language open source software library ( version 2 . 4 . 0 ) . The ANOVA analysis was carried out with a fixed effect permutation ANOVA model consisting of the independent variable treatment ( live Bb versus lysed Bb ) . To identify differentially regulated genes , we performed the F-tests F1 , F2 , F3 and Fs that are implemented in the R/MAANOVA software and assessed gene-centric ( F1-test ) and array centric ( F3 test ) variance through comparison of the respective lists of regulated genes with those provided by the F2 and Fs-tests which interpolate between gene centric and array centric variance . R/MAANOVA uses the Benjamini-Hochberg test [123] to perform the false discovery rate , which was set to a p value of <0 . 005 , and pooled across all gene list to create an inclusive master list and selected the subset of highly significantly differentially regulated genes through filtering for genes that demonstrated a change in their expression level of at least 1 . 3 fold . The differentially regulated genes were organized into clusters using the hierarchical clustering module of D-Chip software ( revised 2006 ) [124] for both genes and samples with the clustering parameter set to ( a ) Euclidian distance and ( b ) a p-value of 0 . 05 . Differentially expressed genes were annotated using all publicly available databases including the GenBank , GO and IHOP databases . Isolated PBMCs were washed 3 times with phosphate buffered saline ( PBS ) , using low speed spins to avoid platelet contamination , then counted using a hemocytometer and resuspended in 30 µl of ice-cold sorting buffer ( PBS , 2 mM EDTA , 0 . 5% BSA , pH 7 . 2 ) per 1×10e7 total cells . Monocytes were isolated from the pelleted PBMCs using a magnetic cell sorting monocyte isolation kit ( Miltenyi Biotech , Auburn , CA ) . PBMCs were first incubated with metallic bead-conjugated antibodies specific for CD3 , CD19 , CD123 and CD56 . Antibody labeled PBMCs were then passed through two separate ferromagnetic columns . The collected flow-through contained CD14+ monocytes ( average purity 96 . 7% ) , which then were counted , spun and resuspended at 1×10e6 per ml in RPMI ( Invitrogen ) containing 10% fetal bovine serum ( FBS ) . Monocytes were plated and incubated for either 4 or 8 hours at 37°C/5% CO2 with live GFP-Bb or equal amounts of lysed spirochetes at multiplicities of infection ( MOIs ) of 1 , 10 , and 100 ( except were noted in the text ) , 100 ng/ml of LPS ( Sigma-Aldrich ) , or 10 µg/ml of Mitogenic Pentapeptide ( MMP ) , a synthetic lipohexapeptide corresponding to the N-terminus of Escherichia coli murine lipoprotein ( Bachem Bioscience , King of Prussia , PA ) . At the end of the incubation period , culture supernatants were collected and store at −70°C for later cytokine analysis . Monocytes also were harvested for flow cytometry , confocal microscopy and/or RNA extraction depending on the experiment . All culture media and reagents utilized in the stimulation experiments were confirmed to be essentially free of LPS contamination ( <10 pg/ml ) by Limulus amoebocyte lysate assay quantification ( Cambrex , MA ) . Up- or down-regulation of selected transcripts generated from the PBMC array was also verified in cDNA from stimulated PBMCs and isolated human monocytes for selected genes by qRT-PCR analysis . RNA was extracted from both stimulated and unstimulated cells using the Paxgene blood RNA kit ( Qiagen , Valencia , CA ) . The quality of the RNA was verified both with the DU 530 , Life Science spectrophotometer ( Beckman , Fullerton , CA ) and Agillent bioanalyzer . Complementary DNA was prepared from extracted RNA samples using a high capacity cDNA RT kit . ( Qiagen , Foster City , CA ) . PCR amplification was performed using 25-µl reaction mixtures which contained 2 . 5 µl of cDNA , 12 . 5 µl of universal master mix ( Applied Biosystems ) , 8 . 75 µl of water and 1 . 25 µl of each primer probe of interest ( 20× ) . Commercially available gene expression assays ( Applied Biosystems ) were used for amplification of the following transcripts; TNF-α ( Hs00174128_m1 ) , IL-1β ( Hs00174097_m1 ) , IL-6 ( Hs00985639_m1 ) , IL10 ( Hs00174086_m1Hs ) , IRF-7 ( Hs00185375_m1 ) ; IFN-β ( Hs00277188_s1 ) , STAT1 ( Hs01014002_m1 ) , USP18 ( Hs00276441_m1 ) , LGALS9 ( Hs00371321_m1 ) , and FXYD6 ( Hs01121135_m1 ) . Internal standard curves for each gene were generated using equal volumes of reverse-transcribed quantitative PCR human reference total RNA ( Clonetech , Mountain View , CA ) diluted 10-fold to obtain concentrations ranging from 125 to 0 . 125 ng/µl . Quantitative RT-PCR gene expression assays for the house keeping gene , glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) ( Hs99999905_m1 ) , were performed using identical aliquots of each cDNA as normalization controls . All amplification reactions were performed in triplicate; control reactions without reverse transcriptase also were performed to confirm the absence of contaminating DNA . Amplification reactions were performed with the iCyclerIQ thermal cycler ( Bio-Rad ) using the following conditions: 95°C for 10 min , and 40 cycles of 95°C for 15 s and 60°C for 1 min . Cycle threshold ( Ct ) values for each gene were determined from the linear region of the corresponding amplification plots using software supplied by the manufacturer ( Bio-Rad ) . Expression levels of all transcripts studied were normalized to the GAPDH level and the relative changes in gene expression generated were calculated using the 2−ΔΔCt method [125] . A 102 µl aliquot of cDNA obtained from unstimulated or stimulated monocytes ( live or lysed Bb at an MOI of 10∶1 ) served as the template for qRT-PCR analysis using a human type I interferon signaling pathway array ( Superarray Bioscience , Frederick MD ) . These arrays contain primer pairs for 84 genes implicated in type I interferon signal transduction as well as housekeeping genes and controls in 96-well microtiter plates . This qRT-PCR methodology directly quantifies transcript levels based upon the 2−ΔΔCT [125] method through measurement of SYBR green fluorescence using nan iQ5 real-time PCR detection system ( Bio-Rad , Hercules , CA ) . The Cytokine Bead Array kit ( BD Biosciences , San Diego , CA ) was used as previously described [2] for simultaneous measurement of tumor necrosis factor alpha ( TNF-α ) , interleukin 6 ( IL-6 ) , IL-10 , IL-1β , and IL-12 in supernatants from stimulated and control human monocytes . A similar kit was used to measure TNF-α , IL-6 and IL-10 in stimulated mouse macrophages . IL-18 was measured by ELISA using a commercial kit ( Bender Med Systems Inc , Burlington , CA ) . Isolated human monocytes or mouse macrophages ( where indicated ) were incubated with 75 nmol/ml LysoTracker Red ( Invitrogen ) plus live Bb-GFP or lysates for a total of four hours ( human monocytes ) or six hours ( mouse macrophages ) . For membrane labeling , isolated cells plus live Bb-GFP were incubated with 5 µg/ml FM4-64 ( Invitrogen ) for 10 minutes on ice . Stimulated cells were fixed in 1% paraformaldehyde , briefly dried onto slides and mounted in Vectashield containing DAPI ( Invitrogen ) . Cells were visualized using epifluorescence or confocal microscopy . Fluorescent images were acquired on an epifluorescent Olympus BX-41 microscope using a 100× ( 1 . 4NA ) oil immersion objective equipped with a Retiga Exi CCD camera ( Q Imaging , Tucson , AZ ) and the following Omega filter sets: DAPI , FITC , and Rhodamine . Confocal images were acquired using a LSM-510 confocal microscope ( Zeiss , Oberkochen , Germany ) equipped with argon and HeNe lasers . Images were acquired using a 63× ( 1 . 3 NA ) oil immersion objective ( 512×512 pixel resolution ) at 1 µm intervals . Image processing and analysis were performed using ImageJ ( NIH , v1 . 41b ) and LSM Image Browser ( Zeiss , v4 . 2 . 0 . 121 ) . The percentage of monocytes or murine macrophages containing intact or degraded fluorescent spirochetes was systematically quantified in an equivalent number of monocytes for each of the conditions studied . All animal work was approved by the UCHC animal care committee and the research conducted according to institutional guidelines . C57Bl/6 mice deficient in TLR2 were kindly provided by Timothy Sellati , PhD . , Albany Medical School . Sex and age-matched wild type ( WT ) C57Bl/6 mice were purchased from Harlan Laboratories . Experiments were done using either mouse peritoneal macrophages ( MPMs ) or bone marrow derived macrophages ( BMDMs ) . General statistical analysis was performed using GraphPad Prism 4 . 0 ( GraphPad Software , San Diego , CA ) . Fold increase or decrease for each specific gene transcript assayed by qRT-PCR and cytokine concentrations were compared amongst the different stimulus by using either a paired or unpaired Student t test or the equivalent non-parametric methods ( i . e . Wilcoxon ) . For each experiment , both the standard deviation and the standard error of the mean were calculated . p values of <0 . 05 were considered significant .
|
Lyme disease is a tick-borne infectious disorder caused by the spirochetal pathogen Borrelia burgdorferi ( Bb ) . Innate immune responses to Bb are thought to be triggered by the spirochete's outer membrane lipoproteins signaling through cell surface toll-like receptors ( TLR1/2 ) . Using a whole genome microarray technique , we showed that live spirochetes elicited a more intense and broader immune response in human peripheral blood mononuclear cells ( PBMCs ) than could be explained simply by TLR1/2 cell surface stimulation . Of particular interest , live Bb also uniquely induced transcription of type I interferons . In similarly stimulated isolated human monocytes , live Bb generated a greater production of pro- and anti-inflammatory cytokines ( TNF-α , IL-6 , IL-10 and IL-1β ) , as well as interferon-β ( IFN-β ) . Secreted IL-18 , which like IL-1β requires cytosolic cleavage of its inactive form by activated caspase-1 , was generated only in response to live Bb . The cytosolic responses occurred despite evidence that phagocytosed spirochetes were rapidly degraded in phagosomal vacuoles , and unable to escape unscathed into the cell cytosol . We conclude that the innate immune signals generated in human monocytes by phagocytosed spirochetes allow the host to control the bacterium through a number of non-exclusive pathways , that are both TLR2-dependent and -independent , and include a type I interferon response .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"infectious",
"diseases/bacterial",
"infections",
"immunology/innate",
"immunity"
] |
2009
|
Activation of Human Monocytes by Live Borrelia burgdorferi Generates TLR2-Dependent and -Independent Responses Which Include Induction of IFN-β
|
Epilepsy is one of the most common neurological disorders in humans with a prevalence of 1% and a lifetime incidence of 3% . Several genes have been identified in rare autosomal dominant and severe sporadic forms of epilepsy , but the genetic cause is unknown in the vast majority of cases . Copy number variants ( CNVs ) are known to play an important role in the genetic etiology of many neurodevelopmental disorders , including intellectual disability ( ID ) , autism , and schizophrenia . Genome-wide studies of copy number variation in epilepsy have not been performed . We have applied whole-genome oligonucleotide array comparative genomic hybridization to a cohort of 517 individuals with various idiopathic , non-lesional epilepsies . We detected one or more rare genic CNVs in 8 . 9% of affected individuals that are not present in 2 , 493 controls; five individuals had two rare CNVs . We identified CNVs in genes previously implicated in other neurodevelopmental disorders , including two deletions in AUTS2 and one deletion in CNTNAP2 . Therefore , our findings indicate that rare CNVs are likely to contribute to a broad range of generalized and focal epilepsies . In addition , we find that 2 . 9% of patients carry deletions at 15q11 . 2 , 15q13 . 3 , or 16p13 . 11 , genomic hotspots previously associated with ID , autism , or schizophrenia . In summary , our findings suggest common etiological factors for seemingly diverse diseases such as ID , autism , schizophrenia , and epilepsy .
Epilepsy is one of the most common neurological disorders in humans with a prevalence of ∼1% and a lifetime incidence of up to 3% [1] . The epilepsies present with a broad range of clinical features , and over 50 distinct epilepsy syndromes are now recognized . Particularly in a pediatric setting , a broad range of different epilepsy syndromes can be distinguished . Seizure disorders can roughly be divided into idiopathic or symptomatic epilepsies . While symptomatic epilepsies are due to an identifiable cause such as metabolic disorders , brain trauma or intracranial tumors , idiopathic seizure disorders occur in the absence of identifiable causal factors and are thought to have a strong genetic contribution . Although it has long been observed that the idiopathic epilepsies have a genetic component , the genetic etiology of only a small fraction of cases can be determined . The role of copy number variants ( CNVs ) in intellectual disability ( ID ) [2]–[8] , autism [9]–[14] and schizophrenia [15]–[19] has been extensively investigated . It has become increasingly clear that , collectively , rare variants contribute significantly to the etiology of these common diseases–following the rare variant common disease hypothesis . We hypothesize this can be extended to other neurological disorders and that rare CNVs significantly contribute to the genetic etiology of epilepsy . Recently , in a study targeted to six genomic regions , recurrent microdeletions on chromosome 15q13 . 3 , 16p13 . 11 and 15q11 . 2 were identified as important genetic factors predisposing to idiopathic generalized epilepsy ( IGE ) [20]–[22] . Here , we carry out whole-genome array comparative genomic hybridization ( CGH ) in a cohort of 517 individuals with mixed types of idiopathic epilepsy in order to discover novel copy number changes associated with epilepsy . We find recurrent microdeletions of 15q13 . 3 , 16p13 . 11 and 15q11 . 2 each in ∼1% of affected individuals , confirming previous studies [20]–[22] . In addition to recurrent rearrangements at rearrangement-prone regions , we show that , overall , 8 . 9% of affected individuals have one or more rare copy number changes involving at least one gene .
We first evaluated rearrangement hotspots for copy number changes . We found 20 probands ( 3 . 9% ) with copy number changes at known rearrangement hotspots including 15q13 . 3 deletions ( n = 5 ) , 16p13 . 11 deletions ( n = 5 ) , 15q11 . 2 BP1–BP2 deletions ( n = 5 ) , 1q21 . 1 deletions ( n = 2 ) , a 16p12 . 1 deletion ( n = 1 ) , a 16p11 . 2 duplication ( n = 1 ) and a more distal 16p11 . 2 deletion ( n = 1 ) ( Table 2 , Figure 1 ) . We also identified four individuals with duplications of 15q11 . 2 BP1–BP2; because duplications of this region are frequent in the general population , we classified these duplications as polymorphic events . These results confirm our previous studies and emphasize the importance of deletions of 15q13 . 3 , 16p13 . 11 and 15q11 . 2 BP1–BP2 as frequent genetic susceptibility factors in epilepsy [20]–[22] . All three regions have also been associated with ID , autism and/or schizophrenia [15] , [17] , [25]–[32] , as have deletions at 1q21 . 1 [33] , [34] , two distinct regions of 16p11 . 2 [10] , [14] , [35]–[37] and 16p12 [38] , which were also detected in our cohort . Deletions of 16p13 . 11 ( 5/517 vs 0/2493 controls , p = 0 . 00014 , Fisher's exact test ) , 15q13 . 3 ( 5/517 vs 0/2493 , p = 0 . 00014 ) and 15q11 . 2 ( 5/517 vs . 4/2493 , p = 0 . 010 ) are significantly enriched in our epilepsy cohort and together account for 2 . 9% of cases . We next focused on non-hotspot CNVs that overlap one or more genes and are not present in the control cohort of 2493 individuals [24] . We identified 28 individuals with at least one rare gene-containing deletion or duplication , and five individuals each carry two rare CNVs ( Table 2 ) . Fifteen of the events we detected involve a single gene . Two genes were altered in two patients each: AUTS2 deletions were identified in one proband with juvenile myoclonic epilepsy ( JME ) and one proband with unclassified non-lesional epilepsy with features of atypical benign partial epilepsy ( ABPE ) [39] . Deletions involving CTYSB ( SPECC1 ) were identified in two probands with IGE . All other single-gene CNVs were seen only once . Seventeen events involved multiple genes , one of which was observed in two different individuals with JME ( duplication of 18q11 , Table 2 ) . We found five individuals with two rare CNVs ( Figure 2 ) . Two patients with JME and a deletion of 16p13 . 11 ( EMJ071 and EMJ117 ) each have a second rare deletion . EMJ071 has a large deletion on chromosome 13 that removes the SLITRK6 gene , a member of the SLITRK gene family involved in controlling neurite outgrowth; individual EMJ117 also has a deletion involving the CTYSB gene . Case ND05260 ( childhood absence epilepsy , CAE ) carries a 647-kb deletion within the GRID2 gene , which encodes a glutamate receptor expressed in the cerebellum , and a 1-Mb duplication of 9q31 . Though both are maternally inherited , neither has been reported in controls . Case EPI 51 ( idiopathic West syndrome ) has two apparently independent duplications of chromosome 5q35 , each containing several genes . Finally , we identified one proband with neonatal convulsions ( NC ) carrying a deletion within the CNTNAP2 gene that spans exons 2–4 as well as a 370-kb deletion of 17p13 involving 7 genes . DNA from one of more family members was available for analysis in 14 cases . Inheritance , if determined , is shown in Table 2 . In twelve cases , we determined that one or both CNVs in the proband were inherited; in three cases the transmitting parent is also affected . In one case ( EP007 . 1 ) , the CNV was not found in the mother , but the father was unavailable . In another case ( K047 ) , parents were unavailable , but a brother was found to carry the same CNV suggesting one of the parents carries the same CNV .
Rearrangements at several genomic hotspots have been associated with a range of neurocognitive disorders . In our cohort of 517 probands with epilepsy , we find deletions at 15q13 . 3 , 16p13 . 11 and 15q11 . 2 in 2 . 9% of our cases . Interestingly , all of the deletions of 15q13 . 3 ( n = 5 ) and 4/5 deletions at 16p13 . 11 and 15q11 . 2 were in probands with IGE , accounting for 3 . 3% of the patients with IGE in our cohort confirming our previous findings . While it is possible that deletions of 15q13 . 3 are also predisposing to non-IGE epilepsy syndromes , we did not find this to be the case in our series ( n = 118 ) . Additional large cohorts of patients with focal epilepsy or epileptic encephalopathy will be required to determine whether these deletions also play a significant role in other subtypes of epilepsy . Deletions of 16p13 . 11 have previously been associated with intellectual disability +/− congenital anomalies in one study [26] . Three of four probands with 16p13 . 11 deletions in that series had epilepsy; two further fetal cases had brain abnormalities . The findings in this cohort and one previous study of IGE [20] suggest that deletions of 16p13 . 11 are more frequent in epilepsy ( 0 . 5–1% of cases ) than in other phenotypes including ID and autism [26] , [27] , [32] , and may be as frequent as 15q13 . 3 deletions in individuals with IGE . Deletions and duplications of this region have also been reported in schizophrenia , though the associations have not been statistically significant [16] , [29] . Deletions of 15q13 . 3 , detected in five individuals with IGE in our series , have been associated with a wide range of phenotypes including ID , autism , epilepsy and schizophrenia [15] , [17] , [20]–[22] , [25] , [28] , [30] , [31] , [40] . The gene within the 15q13 . 3 region that is most likely responsible for the epilepsy phenotype is CHRNA7 , a subunit of the nicotinic acetylcholine receptor . At least two small studies have failed to identify causal point mutations in the CHRNA7 gene in autosomal dominant nocturnal frontal lobe epilepsy [41] and JME [42] , but additional studies should be performed to further evaluate affected individuals for mutations . A recent publication identifying atypical rearrangements with exclusive deletions of CHRNA7 further emphasizes the importance of CHRNA7 as the main candidate gene in this region [43] . Compared to the above structural genomic variants , copy number variation at 15q11 . 2 between breakpoints BP1 and BP2 of the Prader-Willi and Angelman syndrome region is more common in the general population with the BP1–BP2 deletion present in 0 . 2% of unaffected individuals . Despite this , deletions between BP1 and BP2 have now been reported as enriched in patients with schizophrenia [16] , [17] , ID [27] and epilepsy [20] . Furthermore , there is evidence that patients with Prader-Willi or Angelman syndrome who have deletions including BP1–BP2 are more severely affected [44]–[46] . In this study , we also find enrichment of deletions at this locus in affected individuals . Together , these studies suggest that deletion of the 15q11 . 2 BP1–BP2 region confers susceptibility to a wide range of neuropsychiatric conditions , albeit with incomplete penetrance . Two patients in our series , one each with JME and BECTS , have deletions of 1q21 . 1 , which have been previously associated with a wide range of phenotypes , including intellectually disability and developmental delay [33] , [34] , schizophrenia [15] , [17] , [18] , congenital heart disease [47] , [48] and cataracts [34] , [49] . In two large studies of patients who present primarily with cognitive or developmental delay , 5/42 ( 11 . 9% ) patients also had seizures [33] , [34]; 1 of 10 patients with schizophrenia and a 1q21 . 1 deletion also had epilepsy [15] . Identifying 1q21 . 1 microdeletions in patients with idiopathic generalized and idiopathic focal epilepsies suggests that variation at this locus predisposes to a broad range of seizure disorders crossing traditional diagnostic boundaries . In addition , we identified one patient ( EMJ162 ) with JME and a duplication of 16p11 . 2 ( chr16: 29 . 5–30 . 2 Mb ) , which has been associated with autism , developmental delay and schizophrenia [10]–[12] , [14] , [27] , [35] , [37] . Finally , we identified one individual with severe idiopathic generalized epilepsy of infancy ( SIGEI ) ( K027 ) with a more distal deletion of 16p11 . 2 ( chr16: 27 . 7–28 . 9 Mb ) , recently associated with severe early-onset obesity and ID [36] , and one patient with BECTS ( K105 ) and a deletion of 16p12 . 1 ( chr16: 20 . 2–20 . 8 Mb ) , also associated with ID and other neurodevelopmental defects [38] . Thus , our data adds to the phenotypic spectrum associated with rearrangements at several genomic hotspot regions . In particular , we identify hotspot deletions in two patients with BECTS . Gene identification in BECTS , despite representing the most common focal epilepsy syndrome of childhood , has been elusive so far . Here , we suggest that some recurrent hotspot deletions might predispose to both idiopathic generalized and focal epilepsies . We detected 18 deletions and 16 duplications that are not associated with rearrangement hotspots . Fifteen events involve a single gene; of these , 12 are deletions . Although all of the CNVs reported here are not found in our control set of 2493 individuals , it is possible that some are rare but benign CNVs . However , many of the CNVs we identified contain one of more plausible candidate genes for epilepsy ( Table 2 ) . We identified a deletion of exons 2–4 in the CNTNAP2 gene in a proband with neonatal seizures . CNTNAP2 has been identified as a candidate gene for autism [50]–[52] , and heterozygous deletions involving the gene were reported in three patients with schizophrenia and autism [53] . The deletion is predicted to cause an in-frame deletion of 153 amino acids in the resulting protein . The same patient has a 370-kb deletion of 17p13 that deletes seven genes and has not been seen in our control cohort . We also identified a patient with a duplication encompassing a related gene , CNTNAP4 . Finally , two individuals in our cohort have overlapping deletions within AUTS2 . This gene is disrupted by de novo balanced translocations in three unrelated individuals with mental retardation [54] and a pair of twins with autism and mental retardation [55] , suggesting a role for AUTS2 in normal cognitive development . The two deletions we detected are intragenic and overlapping . Previous studies of CNVs in epilepsy have focused on probands with IGE . It is known from studies of families with autosomal dominant epilepsy that a wide range of seizure types can be caused by the same single-gene mutation . For example , Dravet syndrome , a severe early-onset disorder associated with poor cognitive outcome , and the milder generalized epilepsy with febrile seizures plus ( GEFS+ ) syndrome are both caused by mutations in the SCN1A gene [56]–[58] . Therefore , we included probands with common idiopathic focal epilepsies and non-lesional , idiopathic epilepsies . Some of our probands were diagnosed with specific epilepsy syndromes , including myoclonic astatic epilepsy ( Doose Syndrome ) , atypical benign partial epilepsy [39] , Landau-Kleffner syndrome , idiopathic West syndrome , severe idiopathic generalized epilepsy of infancy [59] and benign neonatal or infantile seizures . These particular epilepsy syndromes are usually associated with normal MRI results . We find that 6 . 6% of probands with IGE and 7 . 9% of those with idiopathic focal epilepsy harbor rare CNVs that may underlie their epilepsy phenotype . Notably , 12 . 7% of patients with other , often more severe forms of epilepsy in our series carry one or more rare CNVs . In our series , the vast majority of patients with deletions of 15q13 . 3 , 16p13 . 11 and 15q11 . 2 BP1–BP2 were in the IGE cohort , accounting for 3 . 3% of cases . In the non-IGE patients , a deletion of 15q11 . 2 was found in a single patient with infantile seizures and a deletion of 16p13 . 11 was found in one patient with BECTS , suggesting that deletions at these three genomic hotspots confer greater risk for IGE than other types of epilepsy . In summary , we find that 46/517 probands ( 8 . 9% ) with various forms of idiopathic epilepsy carry one or more rare CNVs that may predispose to seizures , a frequency similar to that in studies of patients who present with other neurocognitive phenotypes , including ID , autism and schizophrenia . Furthermore , we identified CNVs involving genes and genomic regions previously identified in patients with the neurocognitive phenotypes listed above , suggesting common genetic etiological factors for these disorders . Our data suggest that rare CNVs are important in many subtypes of idiopathic epilepsies , including idiopathic generalized and idiopathic focal epilepsies as well as specific idiopathic , non-lesional epilepsy syndromes . The genomic regions and genes identified in this study are potential novel candidate genes for epilepsy .
Patients were collected at five centers after appropriate human subjects approval and informed consent at each site . Patients were collected at five centers: ( 1 ) 140 probands with a primary diagnosis of JME , CAE , absence epilepsy , IGE or idiopathic epilepsy were selected from the NINDS repository ( http://ccr . coriell . org/ninds ) ; ( 2 ) 160 patients are probands with a primary diagnosis of JME from Switzerland . Patients from cohorts ( 1 ) and ( 2 ) were previously analyzed using MLPA for the CHRNA7 gene [60] , and two probands ( EMJ001 and EMJ020 ) were determined to have 15q13 . 3 microdeletions by that method; they were not previously analyzed for any other copy number changes . ( 3 ) 186 German patients came from two cohorts: 76 patients from a population-based cohort from Northern Germany ( POPGEN cohort ) and 110 patients with childhood-onset epilepsy collected at the University of Kiel . Finally , 41 patients with various idiopathic generalized epilepsies collected at ( 4 ) the University of Iowa and ( 5 ) at Washington University , St . Louis . DNA from the NINDS repository was derived from cell lines; DNA from all other cohorts was directly from blood . Patients were diagnosed according to the widely used 1989 ILAE classification [61] . In addition , several pediatric patients were diagnosed with specific syndromes not yet recognized in the ILAE classification ( Table 1 ) . Patients with non-lesional , idiopathic epilepsies in which diagnostic criteria of the recent ILAE classification for particular epilepsy syndromes were not met were labeled as “unclassified” . Array CGH was performed using either custom or commercially available oligonucleotide arrays containing 135 , 000 isothermal probes ( Roche NimbleGen , Inc . ) . Customized arrays ( 459 samples ) were designed with higher density probe coverage in known rearrangement hotspot regions ( average probe spacing 2 . 5 kb ) with lower density whole-genome backbone coverage ( average probe spacing 38 kb ) . A subset of samples ( n = 62 ) was analyzed using a commercially available whole-genome array ( Roche NimbleGen 12×135 k whole-genome tiling array ) with average probe spacing throughout the genome of 21 kb . Data were analyzed according to manufacturer's instructions using NimbleScan software to generate normalized log2 fluorescence intensity ratios . Then , for each sample , normalized log intensity ratios are transformed into z-scores using the chromosome-specific mean and standard deviation . Z-scores are subsequently used to classify probes as “increased” , “normal” and “decreased” copy-number using a three-state Hidden Markov Model ( HMM ) . The HMM was implemented using HMMSeg [62] , which assumes Gaussian emission probabilities . The “increased” and “decreased” states are defined to have the same standard deviation as the “normal” state but with mean z-score two standard deviations above and below the mean , respectively . Probe-by-probe HMM state assignments are merged into segments according to the following criteria: consecutive probes of the same state less than 50 kb apart are merged , and if two segments of the same state are separated by an intervening sequence of ≤5 probes and ≤10 kb , both segments and intervening sequence are called as a single variant . CNV calls are filtered to eliminate ( i ) events containing <5 probes , ( ii ) CNVs with >50% overlap in a series of 2493 control individuals [24] and ( iii ) events that had no overlap with RefSeq genes . In addition , when comparing CNV calls to control CNVs , we eliminated calls for which there was insufficient probe coverage ( <5 probes ) in the control data to identify the same or similar CNV . Filtered copy number changes are also visually inspected in a genome browser .
|
Epilepsy , a common neurological disorder characterized by recurrent seizures , affects up to 3% of the population . In some cases , the epilepsy has a clear cause such as an abnormality in the brain or a head injury . However , in many cases there is no obvious cause . Numerous studies have shown that genetic factors are important in these types of epilepsy , but although several epilepsy genes are known , we can still only identify the genetic cause in a very small fraction of cases . In order to identify new genes that contribute to the genetic causes of epilepsy , we searched the human genome for deletions ( missing copies ) and duplications ( extra copies ) of genes in ∼500 patients with epilepsy that are not found in control individuals . Using this approach , we identified several large deletions that are important in at least 3% of epilepsy cases . Furthermore , we found new candidate genes , some of which are also thought to play a role in other related disorders such as autism and intellectual disability . These genes are candidates for further studies in patients with epilepsy .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"neurological",
"disorders/neurogenetics",
"genetics",
"and",
"genomics",
"neurological",
"disorders/epilepsy"
] |
2010
|
Genome-Wide Copy Number Variation in Epilepsy: Novel Susceptibility Loci in Idiopathic Generalized and Focal Epilepsies
|
The inactivation of p53 creates a major challenge for inducing apoptosis in cancer cells . An attractive strategy is to identify and subsequently target the survival signals in p53 defective cancer cells . Here we uncover a RUNX2-mediated survival signal in p53 defective cancer cells . The inhibition of this signal induces apoptosis in cancer cells but not non-transformed cells . Using the CRISPR technology , we demonstrate that p53 loss enhances the apoptosis caused by RUNX2 knockdown . Mechanistically , RUNX2 provides the survival signal partially through inducing MYC transcription . Cancer cells have high levels of activating histone marks on the MYC locus and concomitant high MYC expression . RUNX2 knockdown decreases the levels of these histone modifications and the recruitment of the Menin/MLL1 ( mixed lineage leukemia 1 ) complex to the MYC locus . Two inhibitors of the Menin/MLL1 complex induce apoptosis in p53 defective cancer cells . Together , we identify a RUNX2-mediated epigenetic mechanism of the survival of p53 defective cancer cells and provide a proof-of-principle that the inhibition of this epigenetic axis is a promising strategy to kill p53 defective cancer cells .
Because activated p53 is a potent inducer of apoptosis [1] , the activation of p53-dependent apoptosis provides an important molecular basis for killing cancer cells . Chemotherapy and radiotherapy , which cause DNA damage , can activate p53 and induce apoptosis in cancer cells . Many cancer cells have amplification of the MDM2 gene , which encodes an E3 ligase of p53 [2] . Thus , compounds that relieve p53 from the inhibition of MDM2 , such as Nutlin and RITA , were sought and discovered [3 , 4] . Compounds that restore specific p53 mutants to the wild type p53 conformation have also been reported [5] . These p53-centric approaches require either the existence of wild type p53 or a specific p53 mutation . However , when p53 is deleted or mutated in other sites , the pro-apoptotic effects of these approaches diminish . Therefore , the loss-of-function of p53 still represents a big challenge for killing p53 defective cancer cells . An attractive alternative approach to killing p53 defective cancer cells is to identify survival signals in cancer cells and subsequently inhibit these survival signals [6] . Preferably , the inhibition of this ( these ) survival signal ( s ) should induce p53-independent apoptosis . Despite many years of genetic studies and recent genome-wide sequencing endeavors , knowledge of these survival signals in p53 defective cells is largely lacking . It is possible that different cancer types have different survival signals in the absence of p53 . One of the cancer types that have high frequency ( ~90% ) of inactivating p53 is osteosarcoma ( OS ) , the most common primary malignant bone tumor in children , adolescents and young adults [7–9] . Thus far , there is no FDA-approved targeted therapy for OS cells . The current standards of care are neoadjuvant chemotherapy followed surgery and adjuvant chemotherapy [10] . The tumor suppressive function of p53 in OS is conserved between human and mouse . Li-Fraumeni syndrome patients who carry p53 mutations have a high risk of developing various cancers including osteosarcoma [11] . Mice with p53 heterozygous deletion develop a high incidence of OS [12] . Thus , OS cells are a good model to study the survival signals of p53 defective cancer cells . The cell-of-origin of OS is currently debatable . Both bone marrow-derived mesenchymal stem cells ( BMSCs or MSCs ) and osteoblasts have been presumed to be the cells-of-origin of OS [13–15] . RUNX2 is a lineage transcription factor for bone development . The 6p21 region that contains the RUNX2 gene is amplified in some osteosarcomas , in consistent with a role of RUNX2 in osteosarcomagenesis [16] . We and others have previously identified RUNX2 ( Runx2 in mouse ) as a p53-repressed target in OS and bone marrow-derived MSCs [17 , 18] . In this study , we explored the roles of RUNX2 in OS cells and our initial hypothesis was that RUNX2 knockdown delays the osteogenic differentiation of OS cells since RUNX2 inhibition delays osteogenic differentiation of MSCs . Unexpectedly , we found that RUNX2 depletion led to the apoptosis of OS cells . The loss of p53 enhances the apoptosis caused by RUNX2 depletion . Using integrative genome-wide approaches , we identified MYC as a novel target of RUNX2 in OS cells . Exogenously expressed MYC partially rescued apoptosis caused by RUNX2 knockdown . Furthermore , we found that RUNX2 recruits an epigenetic complex , the Menin/MLL1 complex , to facilitate the expression of MYC . Inhibition of the Menin/MLL complex using small-molecule inhibitors decreased the expression of MYC and induced apoptosis of p53 defective OS cells .
To examine the levels of RUNX2 in OS cells , we performed immunoblotting ( I . B . ) analysis of RUNX2 in four different human OS cell lines and primary human mesenchymal stem cell ( hMSC ) from two individuals ( Fig 1A ) . SAOS2 and Hu09-M112 cells are p53 null , U2OS cells are p53 wild type , and HOS-MNNG cells carry a R156P p53 mutation [19] . RUNX2 levels were high in SAOS2 and Hu09-M112 cells and low in U2OS , HOS-MNNG cells , and hMSCs ( Fig 1A ) . We then investigated the effect of RUNX2 knockdown on OS cells . We chose SAOS2 and Hu09-M112 cells for our initial study because they express high levels of RUNX2 and undetectable levels of p53 protein ( Fig 1A ) . In contrast to our initial hypothesis that RUNX2 affects OS cell differentiation , we found that RUNX2 knockdown led to apoptosis of SAOS2 cells based on cleaved PARP and propidium iodide staining ( Fig 1B and 1C ) . SAOS2 cells with RUNX2 knockdown eventually died after several passages ( Fig 1D ) . We observed similar results in Hu09-M112 , whereby RUNX2 knockdown increased apoptosis ( S1A–S1D Fig ) . RUNX2 depletion significantly decreased the size of tumors generated from xenografts of SAOS2 cells in immunocompromised NSG mice ( Fig 1E and 1F ) , suggesting that RUNX2 is also required for the survival of OS cells in vivo . Tumors in the absence or presence of RUNX2 knockdown have OS characteristics , such as osteoid formation ( Fig 1G ) . Therefore , we concluded that RUNX2 is required for the survival of p53 null OS cells . Since both SAOS2 and Hu09-M112 are p53 null , one possibility is that the loss of p53 reprograms these cells to depend on RUNX2 to survive . To test whether the apoptosis caused by RUNX2 knockdown requires p53 loss , we performed RUNX2 knockdown in U2OS ( p53 wild type ) and HOS-MNNG ( p53R156P mutant ) cells . RUNX2 knockdown caused apoptosis in both U2OS ( Fig 2A and 2B and S1E Fig ) and HOS-MNNG cells ( Fig 2C and 2D and S1F Fig ) . Thus , RUNX2 depletion leads to apoptosis of OS cells irrespective of p53 status . To examine whether the pro-survival function of RUNX2 exists in hMSCs , we reduced the level of RUNX2 in hMSCs ( Fig 2E ) . RUNX2 knockdown did not have any effects on apoptosis or proliferation of hMSCs ( Fig 2E and 2F and S1G Fig ) . hMSCs had slightly lower levels of RUNX2 as compared to U2OS and HOS-MNNG cells , suggesting that the levels of RUNX2 cannot fully explain the pro-survival function of RUNX2 in OS cells . We then addressed whether the levels of RUNX2 correlate with the degree of apoptosis of OS cells . The protein ( Fig 1A ) and mRNA levels ( Fig 2G ) of RUNX2 were correlated with the degree of apoptosis in OS cells ( Fig 2H ) . RUNX2 knockdown had a more prominent effect on apoptosis in SAOS2 and Hu09-M112 cells than in U2OS and HOS-MNNG cells . To investigate whether the pro-survival function of Runx2 is conserved in mouse OS cells , we reduced the levels of Runx2 in Dunn cells ( a mouse OS cell line ) , mouse MSCs ( mMSCs ) and MC3T3_E1 cells ( a mouse pre-osteoblast cell line ) by knockdown . Runx2 depletion only caused apoptosis in Dunn cells but not in non-transformed mMSCs and MC3T3_E1 ( Fig 2I and 2J ) , suggesting that the pro-survival function of Runx2 is conserved in mouse OS cells . Along with the observation in human OS cells and hMSCs , it appears that RUNX2 acquires a conserved pro-survival function in OS cells while this function is dispensable for the survival of mMSCs and pre-osteoblasts . Our results showed that p53 is not required for RUNX2 depletion-caused apoptosis , since RUNX2 knockdown caused apoptosis in p53 null SAOS2 and Hu09-M112 cells . However , it is unclear whether p53 loss has an effect on the degree of apoptosis in cells expressing wild type p53 . Although a higher degree of apoptosis was observed in p53 null SAOS2 and Hu09-M112 cells than in p53 wild type U2OS cells , the difference could result from genetic difference of the individuals from which these cells are derived . To overcome this genetic background issue , we used CRISPR ( clustered regularly interspaced short palindromic repeats ) technology to generate p53 knockout ( KO ) isogenic cell lines from U2OS cells ( Fig 3A ) [20] . We used two different pairs of CRISPR constructs to delete part of the Exon 4 of the p53 mRNA ( Fig 3A ) , and the deletions caused a loss of p53 protein in U2OS cells ( Fig 3B ) . Using p21 as a functional readout , we confirmed that these clones ( KO_88 and KO_126 ) have loss of function of p53 ( Fig 3B ) . We then performed RUNX2 knockdown in these p53 knockout clones as well as the parental clone ( p53_WT ) and found that the loss of p53 enhances the apoptosis induced by RUNX2 knockdown ( Fig 3C ) . To explore the molecular mechanisms underlying the de novo pro-survival function of RUNX2 in OS cells , we set out to test the possibility that RUNX2 acquires an OS cell-specific binding partner that confers the pro-survival function on RUNX2 . For this , we established a SAOS2 cell line that stably expresses FLAG-tagged RUNX2 ( Fig 4A ) and performed FLAG IP followed by mass spectrometry to detect the proteins enriched in FLAG-RUNX2 IP . We identified CBFB ( core binding factor beta ) as a binding partner of RUNX2 in SAOS2 cells ( Fig 4A and 4B ) and Hu09-M112 cells ( S2A and S2B Fig ) . Reciprocal co-IP experiments showed that endogenous RUNX2 and CBFB interact in OS cells ( Fig 4C ) . CBFB has previously been shown to be a bona fide binding partner of RUNX2 in osteoblasts and to enhance the binding of RUNX2 to DNA [21] . Therefore , CBFB binding to RUNX2 in OS cells cannot explain the de novo survival function of RUNX2 in OS cells . Western blot analysis showed that CBFB is up-regulated in OS cells compared to hMSCs ( Fig 4D ) . Realtime PCR showed that the mRNA levels of CBFB were higher in OS cells compared to hMSCs ( Fig 4E ) . The mechanism underlying the up-regulation of CBFB in OS cells is currently unknown . The degree of the up-regulation does not appear to be p53-dependent . We then tested whether CBFB is also required for the survival of OS cells and found that CBFB depletion in SAOS2 and Hu09-M112 cells gave rise to apoptosis in these two cell lines ( Fig 4F–4I ) . However , CBFB knockdown in hMSCs did not lead to apoptosis ( S2C and S2D Fig ) . Therefore , CBFB knockdown has the same outcome as RUNX2 knockdown . Since CBFB and RUNX2 both are involved in transcriptional regulation , these results suggest the downstream transcriptional targets of RUNX2/CBFB complex mediate the pro-survival function of RUNX2 . To identify the downstream targets of RUNX2 that may be involved in the pro-survival function of RUNX2 , we used an integrative genome-wide approach combining ChIP-seq ( chromatin immunoprecipitation followed by deep sequencing ) and RNA-seq [22] . We performed ChIP-seq of RUNX2 and CBFB in SAOS2 cells . Using two well-established targets of RUNX2 , SP7 ( also known as Osterix ) and ALPL , we found that the binding sites of RUNX2 and CBFB were highly overlapping at these two loci ( Fig 5A ) . CBFB does not directly bind to DNA . Instead , it hetero-dimerizes with the members ( RUNX1 , RUNX2 and RUNX3 ) in the RUNX family and enhances their DNA binding [23 , 24] . At the genome-wide level , 61% ( 6022 out of 9941 ) of RUNX2 peaks overlapped with CBFB peaks ( Fig 5B ) . Peak intensities ( binding strength ) of the 6022 overlapping peaks of RUNX2 and CBFB are highly correlated ( Spearman correlation , cor = 0 . 84 , p<2 . 2e-16 ) , supporting the notion that CBFB facilitates the DNA binding of RUNX2 ( Fig 5C ) . We identified more CBFB peaks than RUNX2 peaks presumably due to the fact that CBFB may act as a transcriptional cofactor for other transcription factors [24 , 25] . The high degree of overlapping of RUNX2 and CBFB binding suggests that the transcriptional regulation by RUNX2/CBFB mediates the pro-survival function of RUNX2 . After obtaining the high quality ChIP-seq dataset for RUNX2 , we performed RNA-seq analyses of SAOS2 cells transduced with shLuc , shRUNX2_3 or shRUNX2_4 . We selected only those transcripts that were changed in the same direction with both shRNAs to minimize off-target effects . These analyses resulted in 209 RUNX2-dependent genes ( S1 Table ) . We assigned RUNX2 peaks from ChIP-seq to specific genes and identified 5770 genes that have at least one RUNX2 binding site less than 25 kb away ( S2 Table , also see Methods ) . Because the number of targets bound by RUNX2 is much greater than that of targets with RUNX2-dependent changes in expression ( Fig 5D ) , only a small fraction of RUNX2 binding sites is functional in transcription . We then integrated RNA-seq data with the ChIP-seq dataset to identify RUNX2 direct targets ( Fig 5D ) , which are defined as those with a RUNX2-dependent change in gene expression ( measured by RNA-seq ) and having at least one associated RUNX2 peak ( measured by ChIP-seq ) . This integration resulted in 112 RUNX2 direct targets: 53 RUNX2-repressed and 59 RUNX2-activated ( Fig 5E ) . To narrow down the critical mediator of the pro-survival function of RUNX2 , we searched for enriched pathways and focused on those related to apoptosis or cell death . A total of 11 apoptosis-related pathways were enriched in RUNX2 direct targets ( Fig 5F ) . We then interrogated the genes within the 11 enriched apoptosis-related pathways and found that MYC is one of the commonly shared genes among the pathways . MYC is known to play important roles in proliferation and survival of many cancer cells [26] . Both RUNX2 and CBFB bound to the promoter of the MYC gene ( Fig 5G ) . Knockdown of RUNX2 and CBFB decreased the expression of the MYC gene in SAOS2 cells ( Fig 5H and 5I ) and Hu09-M112 cells ( S3 Fig ) , further validating our finding that MYC is a downstream target of the RUNX2/CBFB complex in OS cells . Double knockdown of RUNX2 and CBFB in SAOS2 cells still did not lead to a complete loss of MYC expression ( Fig 5J ) , probably due to the incomplete loss of RUNX2 and CBFB in the knockdown or the existence of other regulatory mechanisms of MYC expression . To test whether MYC mediates the pro-survival function of RUNX2 and CBFB , we established SAOS2 cells that stably carry either an empty vector or a MYC-expressing vector . We then reduced RUNX2 levels these SAOS2 cells by knockdown . Exogenously expressed MYC significantly decreased apoptosis caused by RUNX2 depletion ( Fig 6A–6C ) . The decreased apoptosis was also associated with increased cumulative cell number ( S4A Fig ) . We also found that exogenous MYC expression rescued CBFB knockdown ( Fig 6D–6F and S4B Fig ) , supporting the notion that MYC is a mediator of the pro-survival function of the RUNX2/CBFB complex in OS cells . Since exogenous MYC only partially rescues the apoptosis caused by RUNX2 or CBFB knockdown ( Fig 6B and 6E ) , other factors exist to mediate the pro-survival function of the RUNX2/CBFB complex in OS cells . To examine the expression levels of MYC in OS cells , we performed immunoblotting of MYC in hMSCs and OS cells and found that MYC is up-regulated in OS cells compared to hMSCs ( Fig 7A ) . The mRNA of MYC was also significantly higher in OS cells compared to hMSCs with the exception of the HOS-MNNG OS cell line ( Fig 7B ) . We also observed that Myc is up-regulated in mouse OS cells ( S5A Fig ) , suggesting that up-regulation of MYC ( Myc ) is a conserved phenomenon in OS cells . MYC positive staining was also readily observed in the xenografts of SAOS2 and Hu09-M112 cells in NSG mice ( Fig 7C ) and human OS patient samples ( Fig 7D and S5B Fig ) . Knockdown of MYC in SAOS2 cells increased apoptosis ( Fig 7E ) and decreased the cumulative cell number ( Fig 7F ) . Similar results were observed in Hu09-M112 cells ( S5C and S5D Fig ) . Together , our results revealed that MYC is up-regulated by RUNX2 and required for the survival of OS cells ( Fig 7G ) . We explored a possible epigenetic and/or transcriptional regulation of MYC transcription in OS cells since genome-wide sequencing studies did not identify any consistent oncogenes in OS [7 , 8] . Using ChIP-seq , we mapped histone H3 lysine 4 trimethylation ( H3K4me3 ) , a transcription initiation mark , and H3 lysine 79 dimethylation ( H3K79me2 ) , a transcription elongation mark . We found that H3K4me3 and H3K79me2 levels are high in SAOS2 cells compared to hMSCs , supporting an epigenetic and/or transcriptional mechanism for the up-regulation of the MYC gene ( Fig 8A ) . Because RUNX2 regulated the expression of MYC , we investigated whether RUNX2 is involved in the up-regulation of these two histone modifications . We found that RUNX2 knockdown decreased the levels of these two modifications as well as RNA polymerase II ( Pol II ) ( Fig 8B ) . Complexes that carry out H3K4me3 are the COMPASS-like complexes , which consist of MLL1 , Menin and other subunits [27] . Interestingly , both MLL1 and Menin , like RUNX2 , are involved in the normal physiology of skeletal system and osteocyte maturation [28 , 29] . Although previous studies showed that the COMPASS-like complex suppresses leukemogenesis , the tumor promoting role of this complex in solid tumors , such as breast tumors , has been noted [30] , suggesting a contextual role of this complex in cancer . Since Menin is the scaffold protein in the COMPASS complex , we tested the interaction between RUNX2 and Menin and detected interaction between these two proteins at endogenous levels ( Fig 8C ) . To further test whether MLL1 and Menin directly target MYC expression , we performed ChIP assays ( Fig 8D ) . Menin , MLL1 , and another subunit Wdr5 were recruited to the MYC promoter ( Fig 8D ) . Importantly , RUNX2 knockdown decreased the recruitment of these proteins ( Fig 8D ) , suggesting that the recruitment of the COMPASS-like complex to the MYC promoter is regulated by RUNX2 . Similar to RUNX2 knockdown , MLL1 and Menin knockdown decreased the expression of MYC and increased apoptosis ( Fig 8E and 8F ) . The interaction of Menin and MLL1 is required for the activity of MLL1 [31] . Recently , two inhibitors of disrupting the Menin/MLL1 interaction , Mi-2 and Mi-3 ( Menin inhibitor 2 and 3 ) , have been reported [32] . We decided to test whether these two inhibitors induce apoptosis in OS cells . As a control , we included an inactive inhibitor , Mi-nc . Co-IP experiments in SAOS2 cells showed that Mi-2 and Mi-3 decreased the interaction between Menin and MLL1 while Mi-nc did not ( Fig 9A ) . We found that Mi-2 and Mi-3 induced the apoptosis and decreased MYC expression in OS cells ( Fig 9B and 9C ) , suggesting that pharmacological inhibition of Menin/MLL1 interaction induces apoptosis in OS cells . Although we cannot completely exclude the possibility that Mi-2 and Mi-3 affect the expression of other regulators of MYC expression , our results strongly support the notion that the Menin/MLL1 complex is involved in the regulation of MYC expression based on the recruitment of Menin and MLL1 to the MYC promoter ( Fig 8D ) . Together , our data provide a proof-of-concept that the pharmacological inhibition of this epigenetic complex is a promising strategy of inducing apoptosis of OS cells ( Fig 9D ) .
Sarcomas , on average , have less number of mutations compared to other types of tumors , suggesting that epigenetic mechanisms play important roles in their etiology [33] . Here , we studied the pro-survival function of RUNX2 in OS cells and found that RUNX2 induces the expression of MYC . Further , we discovered an epigenetic mechanism involving RUNX2 , the Menin/MLL1 complexes , and MYC that regulates the survival of OS cells ( Fig 9D ) . The Menin/MLL1 complexes have previously been shown to be tumor suppressive in blood cells as the mutations or translocations of MLL are associated with leukemogenesis [27] . In contrast , a recent study showed that these complexes could promote breast tumor growth , suggesting that the roles of the Menin/MLL1 complexes in cancer depend on cancer types [30] . Our result showed that these epigenetic complexes also are involved in the survival of OS cells . Using small-molecule inhibitors , Mi-2 and Mi-3 , we demonstrated that the inhibition of the activities of these complex induced apoptosis of OS cells . Thus , our studies provide a proof-of-concept for killing OS by inhibiting the activity of an epigenetic complex . The loss of p53 activity creates a major challenge for p53-related cancer therapy . To kill p53 deficient cancer cells , several context-dependent approaches were designed: if cancer cells have over-expressed MDM2 , MDM2 inhibitors , such as Nutlin and RITA , can be used to activate p53 [3 , 4] . If cancer cells contain mutant p53 , compounds that convert mutant p53 into the wild type p53-like conformation can be used [5] . However , in cancer cells with p53 deletions , the benefit of these approaches , which requires the existence of either wild type p53 or mutant p53 , is diminished . Therefore , new approaches are needed for inducing apoptosis in cancer cells with a p53 deletion . An alternative approach is to target the survival signals in cancer cells irrespective of p53 status ( p53 wild type , null or mutants ) [6] . Our results showed that RUNX2 signaling is one of the survival signals in OS cells irrespective of p53 status . Because SAOS2 cells were derived from a chemo-resistant OS tumor , our study also provides a much needed means of killing chemo-resistant OS tumor cells [34] . Interestingly , although we started with OS cells carrying p53 deletion , we found that RUNX2 is also required for the survival of OS cells carrying mutant p53 ( HOS-MNNG ) or even wild type p53 gene ( U2OS ) . It is probably due to the fact that the p53 signaling pathway as a whole is dysregulated in tumor versus non-tumor cells . We did observe higher degree of apoptosis in p53 null OS cells ( SAOS2 and Hu09-M112 ) than in p53 wild type ( U2OS ) and p53 mutant ( HOS-MNNG ) OS cells . We note that these cells are derived from different patients and therefore , the difference in apoptosis may be caused by the individual differences of the patients from which these cells are derived . Using CRISPR to generate isogenic p53 knockout cell lines from U2OS , we established that p53 loss does increase the degree of apoptosis caused by RUNX2 depletion . Because RUNX2 is not needed for the survival of hMSCs and pre-osteoblasts , the pro-survival function of RUNX2 signaling in OS cells serves as an unprecedented opportunity to induce p53-independent apoptosis in these tumor cells . RUNX2 is a key lineage factor for osteogenic differentiation and its level needs to be tightly controlled [35] . We and others have shown that p53 indirectly represses RUNX2 in MSCs and OS cells [18] . Thus , when p53 signaling is dysregulated in OS cells , OS cells could “hijack” the pro-survival function of RUNX2 that is normally needed for osteogenic differentiation . MSCs develop into pre-osteoblasts , osteoblasts , and eventually mature osteocytes . It is possible that the RUNX2/CBFB signaling acquires the pro-survival function in a middle stage during this dynamic differentiation process . If this is true , it explains why RUNX2 depletion does not cause apoptosis in human and mouse MSCs , which represent the initial stage of osteogenic differentiation . RUNX2 has been shown to have a negative role in cell cycle progression [36 , 37] . For example , Galindo et al . showed that when being over-expressed , RUNX2 has an anti-proliferative role in MC3T3 ( pre-osteoblast ) and C2C12 ( myoblast ) cells [36] . Interestingly , siRNA-mediated knockdown of RUNX2 has no effect on cell cycle progression in MC3T3 cells . This result is similar to our observation that shRNA-mediated knockdown of RUNX2 has no effect on the survival of hMSCs , mMSCs , and MC3T3 cells . San Martin et al . examined the cell cycle-associated regulation of RUNX2 and CBFB and found that cell cycle regulation of these two proteins becomes aberrant in OS cells [37] . We did attempt to express RUNX2 at a very high level in SAOS2 cells when we established the stable cell line . Under this condition , we fail to obtain stable cell lines , suggesting that a very high level of RUNX2 impairs cell proliferation or survival . We overcame this negative effect of RUNX2 by lowering the titer of lentiviruses and eventually established SAOS2 cells that stably express Flag-tagged RUNX2 ( Fig 4A ) . This preliminary result agrees to the conclusion that RUNX2 can act as a negative regulator of cell cycle . It appears that the level of RUNX2 has to be high enough for RUNX2 to have a negative effect on cell cycle . In addition , as shown by San Martin et al . , RUNX2 level is dysregulated in OS cells , which may have higher tolerance to the RUNX2 level than osteoblasts or other cell types . The majority of our experiments are knockdown . Under this condition , we observed that RUNX2 is required for the survival of OS cells as RUNX2 depletion leads to apoptosis . Therefore , the roles of RUNX2 in cell cycle regulation and survival are cell type- and context-dependent . In this study , we discovered MYC as a novel transcriptional target of RUNX2 . We found that MYC is over-expressed in both human and mouse OS cell lines compared to hMSCs . Because MYC is a survival factor , this RUNX2/MYC axis provides an explanation for the pro-survival function of RUNX2 . Indeed , exogenously expressed MYC partially rescued the apoptosis caused by RUNX2 and CBFB knockdown , further supporting the notion that RUNX2 provides survival signals to OS cells partially through MYC . Since it is challenging to directly target MYC due to its unstructured conformation [38] , RUNX2-mediated regulation of MYC expression offers a much needed framework for developing agents to suppress MYC in p53-defective OS cells . We note that besides RUNX2 , other factors could also regulate the expression of MYC in OS . Nevertheless , RUNX2 is required for the maintenance of MYC expression as RUNX2 knockdown has a prominent role in the reduction of MYC expression in OS cells . Tumor suppressors of OS cells , such as p53 and RB , have been well documented by genetic approaches . The critical roles of these tumor suppressors in OS are underscored by the high frequency of p53 and RB mutations in OS reported by recent genome-wide sequencing studies [7 , 8] . However , oncogenes for OS are less well known . Our study suggests that MYC is one of the oncogenes in OS cells given that it is highly expressed in both human and mouse OS cells . Genetic alterations , such as mutations and translocations , of the MYC gene have been reported in genome-wide sequencing studies [7 , 8] . The low frequency of these genetic alterations suggests that the dysregulation of MYC in OS cells may involve transcriptional and/or epigenetic mechanisms . Indeed , a transcriptional hypothesis is supported by our observations that RUNX2 is required for transcription of MYC in OS cells . Importantly , depletion of RUNX2 decreased MYC expression and increased the apoptosis of OS cells . As targeting MYC is an actively pursued arena , our study merits future testing of effects of emerging MYC-targeting compounds in killing OS cells . Since the expression of RUNX2 is generally restricted to the skeletal system , an alternative strategy could be targeting the RUNX2/CBFB interaction , which might produce less unwanted effects than targeting the ubiquitously expressed MYC .
hMSCs were purchased from ATCC and grown in hMSC medium ( ATCC ) . U2OS , SAOS2 and HOS-MNNG cells were purchased from ATCC and grown in DMEM+10% FBS+antibiotics . Hu09-M112 is a subclone ( generous gift from Dr . Jun Yokota , Biology Division , National Cancer Center Research Institute , Japan ) from Hu09 cells and grown in RPMI1640+10% FBS+antibiotics [39] . mMSCs were isolated from the bone marrow of p53 knockout mice , as previously described [18] . Dunn cells and MC3T3-E1 cells ( ATCC ) were grown in DMEM+10% FBS+antibiotics . Ten million SAOS2 cells were transduced with lentivirus expressing shLuc , shRUNX2_3 or shRUNX2_4 for 24 hours . Cells were re-suspended in 100 ul of PBS buffer+25mM HEPES , and mixed with 50 ul Matrigel before being transplanted into hind limb muscle of NSG mice . 43 days after transplantation , tumors were harvested , weighed and fixed in 10% neutral buffered formalin for 16 hours before Hematoxylin and Eosin staining . For Hu09-M112 cells , 1 million cells were transplanted into NSG mice . 69 days after transplantation , tumors were harvested , weighed and fixed in 10% neutral buffered formalin for 16 hours before Hematoxylin and Eosin staining . Mice were maintained under the strict guidelines of the Institutional Animal Care and Use Committee ( IACUC ) -approved protocols of the National Cancer Institute and National Heart , Lung , and Blood Institute . To calculate the cumulative cell numbers , cells were split at a constant ratio ( splitting ratio ) for each passage . At each splitting , the cell number was counted and multiplied by the splitting ratio . The final number is the cumulative cell number . The final curve was generated by graphing the cumulative cell numbers over several passages . Note that for each cell line , a no-virus transduction control was included to determine the 0-day of counting , which is the first day when the no-virus control was completely dead after drug selection . ChIP-seq was performed in the next generation sequencing facility at the National Cancer Institute ( NCI ) , as previously described [18 , 22] . Peaks were identified by the MACS algorithm [40] . For calculating peak intensity , the number of tags at each nucleotide within a peak was calculated and summed up for all the nucleotides across the peak . The sum was defined as peak intensity , which was used to calculate the correlation between CBFB and RUNX2 binding . To assign the peak to specific genes , we used the approach we developed previously [22] . Briefly , we arbitrarily defined the promoter region as the region between 5 kb upstream to 5 kb downstream of the transcription start site ( TSS ) of a transcript . The rest of the region within the gene body ( 5 kb downstream of TSS to the end of transcription ) is defined as the gene body region . We also define regions 25 kb away from the transcript as distal . ChIP assay were done in the same way as ChIP-seq . The amplicon enrichment was measured by realtime PCR and calculated as percentage of input . Primers for amplifying the location on the MYC gene are: forward , 5’-ACTCACAGGACAAGGATGCG-3’; Reverse , 5’-TGCTCCTCCGTAGCAGTACT-3’ . SAOS2 cells were transduced with lentiviruses expressing shLuc , shRUNX2_3 or shRUNX2_4 for 4 days when we observed significant RUNX2 protein reduction but no obvious apoptosis . RNA-seq was performed as previously described [18] . We reasoned that at this time point , the underlying transcriptional changes precede the cellular event—apoptosis . After RNA extraction with Trizol , 1ug total RNA was subjected to deep sequencing in the NextSeq 500 machine at the NCI next generation sequencing facility . Reads were aligned to the human genome ( Build hg19 ) . The cufflinks algorithm was used to calculate reads per kilobase per million ( RPKM ) for each RefSeq transcript . shRNAs were cloned into a pLKO . 1 backbone , which carries a puromycin resistant gene . For the rescue experiment , SAOS2 cells were transduced with a plenti6-GW-MYC vector and selected in 10 ug/ml of Blasticidin . Stable cells were then transduced with a RUNX2 or CBFB shRNA vector . shRNA sequences are as follows: RUNX2 ( h ) _3 , top: 5’-CCGGTGCACTATCCAGCCACCTTTACTCGAGTAAAGGTGGCTGGATAGTGCATTTTT-3’ RUNX2 ( h ) _3 , bottom: 5’-AATTAAAAATGCACTATCCAGCCACCTTTACTCGAGTAAAGGTGGCTGGATAGTGCA-3’ RUNX2 ( h ) _4 , top: 5’-CCGGGCTACCTATCACAGAGCAATTCTCGAGAATTGCTCTGTGATAGGTAGCTTTTT-3’ RUNX2 ( h ) _4 , bottom: 5’-AATTAAAAAGCTACCTATCACAGAGCAATTCTCGAGAATTGCTCTGTGATAGGTAGC-3’ RUNX2 ( m ) _4 , top: 5’-CCGGGCAGAATGGATGAGTCTGTTTCTCGAGAAACAGACTCATCCATTCTGCTTTTT-3’ RUNX2 ( m ) _4 , bottom: 5’-AATTAAAAAGCAGAATGGATGAGTCTGTTTCTCGAGAAACAGACTCATCCATTCTGC-3’ RUNX2 ( m ) _5 , top: 5’-CCGGCCGAGTCATTTAAGGCTGCAACTCGAGTTGCAGCCTTAAATGACTCGGTTTTT-3’ RUNX2 ( m ) _5 , bottom: 5’-AATTAAAAACCGAGTCATTTAAGGCTGCAACTCGAGTTGCAGCCTTAAATGACTCGG-3’ CBFB ( h ) _1 , top: 5’- CCGGGAGAAGCAGGCAAGGTATATTCTCGAGAATATACCTTGCCTGCTTCTCTTTTT -3’ CBFB ( h ) _1 , bottom: 5’-AATTAAAAAGAGAAGCAGGCAAGGTATATTCTCGAGAATATACCTTGCCTGCTTCTC-3’ CBFB ( h ) _2 , top: 5’-CCGGCCGCGAGTGTGAGATTAAGTACTCGAGTACTTAATCTCACACTCGCGGTTTTT -3’ CBFB ( h ) _2 , bottom: 5’-AATTAAAAACCGCGAGTGTGAGATTAAGTACTCGAGTACTTAATCTCACACTCGCGG-3’ MLL1 ( h ) _1 , top: 5’-CCGGGATTCGAACACCCAGTTATTCCTCGAGGAATAACTGGGTGTTCGAATCTTTTT -3’ MLL1 ( h ) _1 , bottom: 5’-AATTAAAAAGATTCGAACACCCAGTTATTCCTCGAGGAATAACTGGGTGTTCGAATC-3’ MLL1 ( h ) _5 , top: 5’-CCGGTGCCTGGAAGGAGCCTATTATCTCGAGATAATAGGCTCCTTCCAGGCATTTTT-3’ MLL1 ( h ) _5 , bottom: 5’-AATTAAAAATGCCTGGAAGGAGCCTATTATCTCGAGATAATAGGCTCCTTCCAGGCA-3’ Menin ( h ) _2 , top: 5’-CCGGCTGTACCTGAAAGGATCATACCTCGAGGTATGATCCTTTCAGGTACAGTTTTT-3’ Menin ( h ) _2 , bottom: 5’-AATTAAAAACTGTACCTGAAAGGATCATACCTCGAGGTATGATCCTTTCAGGTACAG -3’ Menin ( h ) _5 , top: 5’- CCGGTCTACGACGGCATCTGCAAATCTCGAGATTTGCAGATGCCGTCGTAGATTTTT -3’ Menin ( h ) _5 , bottom: 5’-AATTAAAAATCTACGACGGCATCTGCAAATCTCGAGATTTGCAGATGCCGTCGTAGA-3’ MYC ( h ) _4 , top: 5’-CCGGCCTGAGACAGATCAGCAACAACTCGAGTTGTTGCTGATCTGTCTCAGGTTTTT-3’ MYC ( h ) _4 , bottom: 5’-AATTAAAAACCTGAGACAGATCAGCAACAACTCGAGTTGTTGCTGATCTGTCTCAGG-3’ MYC ( h ) _5 , top: 5’-CCGGCCTGAGACAGATCAGCAACAACTCGAGTTGTTGCTGATCTGTCTCAGGTTTTT-3’ MYC ( h ) _5 , bottom: 5’-AATTAAAAACCTGAGACAGATCAGCAACAACTCGAGTTGTTGCTGATCTGTCTCAGG-3’ CRISPR targeting sequences ( shown in Figure ) were cloned into the pX330 vector ( Addgene: #42230 ) . To delete a region in the Exon 4 of the human p53 gene , a pair of CRISPR constructs was co-transfected into the cells together with an EGFP expression vector . EGFP-positive cells were selected by flow cytometry and plated at a single cell density . Colonies were picked , propagated , and genotyped by PCR . We used the following antibodies for immunoblotting: RUNX2 ( Cell Signaling , Cat:8486 ) , p53 ( DO1 , Santa Cruz , Cat:sc-126 ) , β-actin ( Sigma , Cat:A5316 ) , Cleaved PARP ( Promega , Cat: G7341 ) , Tubulin ( Sigma , Cat: T3526 ) , CBFB ( Cell Signaling , Cat: 12902s ) , and MYC ( Epitomics , Cat: 1472–1 ) . We used the following antibodies for ChIP: RUNX2 ( house made using recombinant RUNX2 fragments ) , Menin ( Bethyl , Cat: A300-105A ) , CBFB ( Bethyl , Cat: A303-549A ) , RNA polymerase II , N20 ( Santa Cruz , Cat: sc-899 ) , H3K4me3 ( Abcam , Cat: ab8580-100 ) , H3K79me2 ( Abcam , Cat: ab3594-100 ) . MYC immunohistochemistry was performed with a mouse monoclonal antibody of MYC , 9E11 ( Santa Cruz , sc-47694 ) . Mi-2 and Mi-3 were purchased from Selleckchem , and Mi-nc from Cayman Chemical . All the inhibitors were dissolved in DMSO to make 50 mM stock solution . Working concentrations were 50 uM . We established a stable cell line for SAOS2 and Hu09-M112 cells by transducing them with the plenti6-GW-FLAG-RUNX2 vector and selecting transduced cells with 10 ug/ml of Blasticidin . After the stable cell line was established , cells from ten to twenty 10cm plates ( around 60 million cells ) were used to perform a FLAG pulldown . Briefly , cell pellets were lysed in 10ml of NET buffer ( 50 mM Tris , pH 7 . 5 , 250 mM NaCl , 5 mM EDTA , 0 . 1% NP40 , 10% Glycerol with freshly added protease inhibitors ) . Clear cell lysate was incubated with 100ul of anti-FLAG M2 affinity gel ( 200ul slurry ) ( A2220 , Sigma ) overnight at 4°C . FLAG_RUNX2 ( h ) bound anti-FLAG M2 affinity gel was washed three times with NET buffer ( 50mM Tris , pH 7 . 5 , 250mM NaCl , 5mM EDTA , 1% NP40 , 10% Glycerol , freshly added proteinase inhibitors ) and eluted four times with 500ug/ml 3xFLAG peptide ( F4799 , Sigma ) . Combined eluted materials were concentrated with acetone and resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) and silver staining ( SilverQuest Staining Kit , LC6070 , Invitrogen ) . Silver stained bands were cut out from the SDS-PAGE gel and analyzed by Mass Spectrometry . Formalin fixed paraffin embedded slides were deparaffinized in Xylene , 100% ethanol and 95% ethanol . Antigens were retrieved by boiling slides in 10 mM sodium citrate for 10 minutes . After cooling , slides were treated with 3% H2O2 for 10 minutes followed by PBS+0 . 1% Tween 20 washing and blocked with serum . Slides were then incubated with 1:200 MYC antibody ( Epitomics ) for 1 hour at room temperature , washed three times with PBS , incubated with biotinylated goat anti-rabbit IgG secondary antibody ( VECTASTAIN ABC Kit ) for 1 hour at room temperature . After washing with biotin-avidin solution for 30 minutes at room temperature , slides were rinsed with PBS three times , DAB solution was added to allow color development for 2–5 minutes . Eighty eight OS patient tumors were collected under an Institutional Review Board ( IRB ) approved protocol of the Montefiore Medical Center . A TMA was generated using these 88 tumors . Immunohistochemistry ( IHC ) procedures of using this TMA were approved under an exemption of IRB ( #12806 ) by the Office of Human Subjects of Research ( OHSR ) at the National Institutes of Health . Formalin fixed paraffin embedded slides were deparaffinized in Xylene , 100% ethanol and 95% ethanol . Masson’s trichrome staining was performed according to the instructions of the NovaUltra Masson Trichrome Stain Kit ( IHCWorld , IW-3006 ) . Blue color indicates collagen , black color the nuclei , and red color the cytoplasm .
|
Because activated p53 is a potent inducer of apoptosis , several approaches centering on p53 activation are designed for killing cancer cells . However , more than half of human tumors have p53 inactivation , which renders these p53-activating approaches less effective in killing cancer cells . Targeting the survival signals specific to p53 defective cancer cells offers an opportunity to circumvent the challenge of p53 inactivation . In this study , we showed that one such survival signal is the RUNX2 signaling pathway . To investigate the mechanism underlying this survival signal , we used biochemical , genetic , and genomic approaches . The MYC gene was identified as a novel mediator of the pro-survival function of RUNX2 . We further studied the regulatory mechanism of MYC by RUNX2 and found that RUNX2 recruits the Menin/MLL1 epigenetic complex to induce the expression of MYC . Using small molecule inhibitors of the Menin/MLL1 complex , we showed that targeting RUNX2/Menin/MLL1/MYC axis is a feasible strategy for killing p53 defective cancer cells . Our study paves the road for the future development of targeted therapies for OS .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"biotechnology",
"genome",
"engineering",
"cell",
"death",
"engineering",
"and",
"technology",
"cell",
"cycle",
"and",
"cell",
"division",
"synthetic",
"biology",
"cell",
"processes",
"synthetic",
"bioengineering",
"crispr",
"epigenetics",
"tubulins",
"immunologic",
"techniques",
"synthetic",
"genomics",
"research",
"and",
"analysis",
"methods",
"bioengineering",
"synthetic",
"genome",
"editing",
"genomic",
"signal",
"processing",
"proteins",
"gene",
"expression",
"biochemistry",
"signal",
"transduction",
"cytoskeletal",
"proteins",
"immunohistochemistry",
"techniques",
"cell",
"biology",
"apoptosis",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"histochemistry",
"and",
"cytochemistry",
"techniques"
] |
2016
|
A RUNX2-Mediated Epigenetic Regulation of the Survival of p53 Defective Cancer Cells
|
As an obligatory pathogen , influenza virus co-opts host cell machinery to harbor infection and to produce progeny viruses . In order to characterize the virus-host cell interactions , several genome-wide siRNA screens and proteomic analyses have been performed recently to identify host factors involved in influenza virus infection . CD81 has emerged as one of the top candidates in two siRNA screens and one proteomic study . The exact role played by CD81 in influenza infection , however , has not been elucidated thus far . In this work , we examined the effect of CD81 depletion on the major steps of the influenza infection . We found that CD81 primarily affected virus infection at two stages: viral uncoating during entry and virus budding . CD81 marked a specific endosomal population and about half of the fused influenza virus particles underwent fusion within the CD81-positive endosomes . Depletion of CD81 resulted in a substantial defect in viral fusion and infection . During virus assembly , CD81 was recruited to virus budding site on the plasma membrane , and in particular , to specific sub-viral locations . For spherical and slightly elongated influenza virus , CD81 was localized at both the growing tip and the budding neck of the progeny viruses . CD81 knockdown led to a budding defect and resulted in elongated budding virions with a higher propensity to remain attached to the plasma membrane . Progeny virus production was markedly reduced in CD81-knockdown cells even when the uncoating defect was compensated . In filamentous virus , CD81 was distributed at multiple sites along the viral filament . Taken together , these results demonstrate important roles of CD81 in both entry and budding stages of the influenza infection cycle .
Influenza virus , the major causal agent of flu , is an enveloped , negative-sense RNA virus containing three viral membrane proteins: hemagglutinin ( HA ) , neuraminidase ( NA ) , and M2 proton channel . Encapsulated within the viral envelope is a layer of matrix protein ( M1 ) and a segmented genome . The eight single-stranded RNAs package into viral ribonucleoprotein complexes ( vRNPs ) , each attached to a RNA-dependent RNA polymerase complex with three subunits: PA , PB1 , and PB2 [1] . As an obligatory pathogen that encodes only 13 viral proteins , influenza virus must rely on host proteins and cellular machinery to complete its infection cycle . Influenza infection begins with virus binding to sialic acids on the plasma membrane [2] . Virus-receptor interaction subsequently triggers viral entry through multiple endocytic routes including clathrin-mediated endocytosis , a clathrin/caveolin-independent pathway , and macropinocytosis [3]–[7] . Upon internalization , virus particles are trafficked from early endosomes to maturing endosomes , where fusion between the virus and endosomal membranes results in release of vRNPs into the cytoplasm followed by nuclear import of the vRNPs [1] , [8]–[10] . As replication proceeds , viral mRNAs are exported out of the nucleus for protein translation , and viral components are trafficked to the plasma membrane , the site of virus assembly and progeny virion egress [11] . Recently , several genome-wide siRNA screens identified host factors exploited by influenza virus [12]–[16] . CD81 emerged as a top candidate in two screens and was found to regulate early viral entry steps [15] , [16] . CD81 belongs to the family of tetraspanins and is expressed on both plasma and endosomal membranes [17]–[19] . It associates with other tetraspanins and tetraspanin-interacting proteins to form tetraspanin-enriched microdomains [18] , [20] , [21] . Together , these proteins regulate many cellular processes such as cell adhesion , cell signaling , cell migration , and protein trafficking [18] , [19] , [22]–[24] . Tetraspanins are known to play an important role in different steps of viral infection [25] . For example , CD81 functions as a co-receptor for hepatitis C virus ( HCV ) [26]–[29] . CD81 interacts with HCV glycoprotein E2 to prime the virus for low-pH dependent fusion during entry [26] , [30] , [31] . In addition to mediating viral entry , CD81 is also potentially involved in viral assembly . CD81 is one of the cell-derived components incorporated into purified influenza virus particles [32] . It is however unknown how CD81 facilitates influenza viral entry , or whether CD81 plays a functional role in influenza virus assembly . We conducted a comprehensive study from viral entry to egress to examine the effect of CD81 depletion on influenza infection . Upon dissecting each of the major steps in influenza infection pathway , we found that CD81 was required for productive viral infection and that CD81 primarily functions at two stages: viral fusion within endosomes and virus budding . About half of the influenza virus particles that fused in cells underwent fusion within CD81+ endosomes , and CD81 depletion led to a decrease in virus fusion and infection . CD81 was highly enriched in the virus budding zones and recruited to specific sub-viral locations . During virus assembly , CD81 initially formed small clusters at the growing tip of assembling virus , and then localized at both the growing tip and budding neck of spherical or slightly elongated virions . CD81 depletion led to an increase in the propensity of budding virions to remain attached to the plasma membrane and a reduction in progeny virus production . These findings demonstrate a dual function of CD81 in both entry and budding of influenza viruses .
To elucidate the role of CD81 in influenza virus infection , we used CD81 knockdown by siRNA to probe the effect on three different influenza A virus strains: influenza A/WSN/33 ( H1N1 ) , a lab-adapted strain that mainly produces spherical virus particles; influenza A X-31 , A/Aichi/68 ( H3N2 ) which has a slightly elongated shape [33]; and influenza A/Udorn/72 ( H3N2 ) , which can produce long filamentous virus [34]–[37] . We screened six CD81 siRNA constructs , including several previously reported ones [15] , [16] , and found that CD81 siRNA 1 gave the highest knockdown efficiency ( Figure 1A , B and Figure S1A , B ) . After 48 hours , siRNA 1 yielded 80–85% CD81 knockdown ( Figure S1C ) . Treatment with siRNA 1 specifically depleted CD81 , whereas the expression levels of several known CD81-associating proteins , including CD9 , CD82 , CD63 , integrin β1 ( ITGB1 ) , EGFR , and EWIF , were not affected ( Figure S1D ) . In subsequent experiments , we used siRNA 1 to deplete CD81 in A549 lung carcinoma cells . To assay the effect of CD81 depletion on the production of infectious viral progeny , siRNA-treated A549 cells were infected with WSN , X-31 or Udorn virus at a MOI of <0 . 1 for 36 hours . The virus titer of the supernatant was determined using plaque assays . Compared to the non-targeting control siRNA treated cells , the CD81-knockdown cells exhibited a substantial decrease in virus titer: ∼90% decrease for WSN , ∼75% decrease for X-31 , and ∼70% decrease for Udorn ( Figure 1C ) . These results are consistent with the previously published data [15] , [16] , and indicate that multiple different influenza strains require CD81 for infection . Next , we proceeded to determine which stage ( s ) of the multi-step influenza-infection process is CD81-dependent . To test whether CD81 affects early infection , we infected siRNA-treated A549 cells and measured the expression of NP , the first viral protein expressed in influenza-infected cells [15] , [16] , [38] . We assayed the fraction of cells that express NP ( NP+ ) as well as the level of NP expression in each NP+ cell using flow cytometry . For all three viral strains , CD81-knockdown cells had ∼50% fewer NP+ cells , as compared to non-targeting siRNA treated control cells ( Figure 1D ) . Among the NP+ cells , the NP expression level was similar between control and CD81-knockdown cells ( data not shown ) . These results suggest that CD81 is involved in early infection either at the step of or prior to viral protein expression . In order to test whether CD81 directly affects viral protein expression , we next induced viral fusion ( uncoating ) at the plasma membrane through an acid-bypass treatment: treatment with a buffer of pH below 5 , the pH value required for HA-induced membrane fusion . We then probed the expression of NP in these samples . The fraction of NP+ cells and the level of NP expression were similar between control and CD81-knockdown cells infected by the virus using the acid-bypass treatment ( Figure 1E and data not shown ) . These results suggest that CD81 is not directly involved in the viral protein expression , and the inhibition of virus infection by CD81 knockdown was most likely due to inhibition of viral uncoating in endosomes or any step prior to uncoating . It is worth noting that this acid-bypass assay overcame the entry defect for the WSN and X-31 strain , but a similar acid bypass treatment did not work for the Udorn strain , likely because low pH causes fragmentation of filamentous influenza viruses [7] . To probe whether CD81 plays additional roles beyond viral uncoating , we induced influenza viral fusion at the plasma membrane using the acid-bypass treatment to overcome the entry defect , and determined the virus titer of the supernatant 17 hours post-infection . Notably , as compared to control siRNA treated cells , the virus titer was decreased by more than 50% in the CD81-knockdown cells ( Figure 1F ) . This defect did not result from a decrease in viral gene expression , as the percent of viral protein expressing cells and the expression level at 17 hours post-infection were unaffected upon CD81 knockdown ( Figure 1E and data shown later ) . These results suggest that CD81 affects another step post viral gene expression . Taken together , the above results indicate that CD81 is important for two distinct stages of the influenza infection cycle: one during the early infection at or prior to viral uncoating and one during late infection after viral gene expression . In the following experiments , we aimed to identify the specific roles of CD81 in influenza virus infection . To identify the CD81-dependent entry step ( s ) , we conducted a series of experiments to examine the effect of CD81 on virus binding , internalization , transport into early endosomes , and fusion . siRNA-treated A549 cells were first incubated with fluorescently labeled X-31 virus for 30 minutes at 4°C , a temperature that inhibits endocytosis . The amount of surface-bound virus was analyzed through flow cytometry . CD81-knockdown cells exhibited no defect in binding with influenza viruses when compared to control cells treated by non-targeting siRNA ( Figure 2A ) . Next , influenza virus infection was allowed to proceed for 30 minutes at 37°C , and the number of internalized virus particles was quantified . As shown in Figure 2B , the number of internalized viruses per cell was similar between control and CD81-knockdown cells . Moreover , virus particles in both control and CD81-knockdown cells were delivered into early endosomes after internalization at 37°C . At 20 minutes post infection , the mean fraction of virus particles colocalized with early endosomes ( marked by EEA1 ) was ∼50% and ∼53% for control and CD81-knockdown cells , respectively ( Figure 2C ) . These results suggest that CD81 does not affect trafficking of influenza virus to early endosomes . Additionally , we tested the effect of CD81 depletion on two other viruses: respiratory syncytial virus ( RSV ) and GFP-encoding pseudotyped murine leukaemia virus ( MLV ) , which are known to undergo virus fusion in early endosomes in a pH-independent manner [39] , [40] . Pseudotyped MLV infected cells express GFP , but cannot produce complete virions , allowing the quantification of MLV entry through measuring the GFP expression . For RSV , we measured the expression level of the fusion protein ( F protein ) after 24 hours of infection . As shown in Figure 2D , the fraction of infected cells was similar between control and CD81-knockdown cells with MLV and RSV infection . This data further corroborates the notion that CD81 does not affect virus trafficking into early endosomes . Taken together , the results presented above indicate that CD81 is not involved in influenza virus binding , internalization , or trafficking into early endosomes . Next , we probed the role of CD81 in virus fusion . Influenza virus is trafficked from early endosomes to maturing endosomes , in which the low pH environment triggers conformational changes in HA that mediate viral fusion with the endosomal membrane [2] , [9] , [41] . In addition to being distributed on the plasma membrane , CD81 also showed substantial colocalization with early and maturing endosomes , which are Rab5 positive ( Rab5+ ) ( Figure 3A ) [41] . About 30–35% Rab5+ endosomes contained CD81 ( Figure 3A ) , suggesting that CD81 is enriched in a sub-population of these endosomes . To probe whether influenza virus particles are delivered into CD81+ endosomes , we allowed Alexa Fluor 647-labeled X-31 to internalize for 15 minutes and immunostained the cells for CD81 . As shown in Figure 3B , a substantial fraction ( ∼54% ) of virus colocalized with CD81+ endosomes . We then tracked individual influenza virus particles in living cells , a technique that has been previously established [5] , [10] , [41]–[43] , to examine whether influenza viruses fuse in CD81+ endosomes . To this end , we expressed CD81-mEmerald in A549 cells . Similar to endogenous CD81 , CD81-mEmerald was localized on both plasma and endosomal membranes ( Figure S2A ) and the expression of CD81-mEmerald did not affect the fraction of endosomes that are CD81+ ( Figure S2B–D ) . Moreover , the expression of CD81 also did not affect influenza viral fusion or infectivity ( Figure S2E , F ) . To facilitate tracking of individual virus particles , X-31 viruses were labeled with a saturating amount of DiD , a lipophilic dye , such that the fluorescence emission from the DiD molecules was low due to a self-quenching effect between neighboring dyes . Fusion between the virus envelope and endosomal membrane should lead to an increase in fluorescent intensity ( dequenching ) , due to the diffusion of dyes from the virus into the lipid bilayer of the endosomes [42] . We added labeled viruses to the CD81-mEmerald expressing cells in situ at 37°C . The virus particles typically show restricted movement immediately after binding to the cell , followed by rapid and directed movement once the virus particles are internalized , similar to our previous observations [5] , [41] , [42] . We observed a proportion of virus particles entering into CD81+ endosomes soon after internalization , as illustrated by the example shown in Figure 3C and movie S1 . These viruses remained colocalized with CD81 and eventually fused with the CD81+ endosomes , as reflected by the sudden increase of DiD fluorescence , presumably after the endosomes matured to acquire a sufficiently low pH ( Figure 3C and movie S1 ) . Among the 61 virus particles that we tracked from binding to fusion , about 52±8% underwent viral fusion within CD81+ endosomes ( Figure 3E ) . The remaining 48±8% of virus particles fused in endosomes lacking CD81 ( Figures 3D , E , and movie S2 ) . To confirm that the fusion events indeed occurred in endosomes , we tracked individual DiD-labeled influenza virus particles in cells expressing RFP-Rab5 . Similar to previous observations [41] , nearly 90% of the viral fusion events occurred in Rab5+ endosomes ( Figure S3 ) . To investigate whether CD81 affects viral fusion , we next monitored the DiD fluorescent intensity in control and CD81-knockdown cells that were infected with DiD-labeled X-31 virus . In these experiments , cells were first incubated with DiD-labeled virus at 4°C and then the temperature was increased to 37°C to initiate viral entry . At specific time points after the temperature shift , infected cells were collected and the DiD fluorescence from these cells was quantified with flow cytometry . As expected , there was a consistent increase of DiD fluorescence with time due to viral fusion ( Figure 3F ) . Notably , compared to control cells , CD81-knockdown cells exhibited a significant reduction in viral fusion ( Figure 3F ) , suggesting that CD81 facilitates viral fusion . Most of the remaining viral fusion events in CD81-knockdown cells still occurred in Rab5+ endosomes ( Figure S3 ) . The reduction in viral fusion was , however , incomplete ( Figure 3F ) , consistent with the observation that only about half of the virus particles fuse in CD81+ endosomes ( Figure 3E ) , though the incomplete inhibition of viral fusion could also be in part due to the incomplete knockdown of CD81 ( Figure S1 ) . These data indicate that CD81 marks a specific population of Rab5+ endosomes that are responsible for half of viral fusion events . Because CD81-knockdown cells reduced viral infection ( Figure 1C ) and exhibited a higher reduction in virus titer when infected without acid bypass than when infected with acid bypass ( Figures 1C , F ) , viral fusion within CD81+ endosomes likely leads to productive influenza infection . Taken together , our results indicate that half of virus particles are trafficked to and undergo viral fusion in CD81+ endosomes . CD81 could facilitate viral fusion by organizing endosomal membrane to assist viral fusion or helping virus traffick to fusion-competent endosomal compartments . In the subsequent experiments , we aimed to determine which post-entry step ( s ) of the viral infection process are CD81 dependent . To this end , we examined the effect of CD81 on viral protein expression , viral protein trafficking , and the assembly and egress of progeny viruses . Our initial results in Figure 1E suggested that CD81 knockdown did not directly affect viral NP expression . This was further substantiated by infecting cells with various viral doses across different time points using the acid-bypass treatment . As shown in Figure 4A , the fraction of NP+ cells and NP expression level increased with the viral dose and infection time , while there was no significant difference between control and CD81-knockdown cells . To further validate the finding , we measured the expression of another cytosolic viral protein , M1 . Similar to the results on NP , CD81-knockdown cells infected by influenza viruses using the acid-bypass treatment did not exhibit a difference in the fraction of M1+ cells or the level of M1 protein expression , when compared to the control cells ( Figure 4B ) . These data suggest that CD81 does not play a role in viral gene expression for cytosolic viral proteins . To determine whether CD81 affects viral membrane protein expression , we probed M2 expression with acid bypass treatment . The fraction of M2+ cells and the M2 expression level in M2+ cells were similar between control and CD81-knockdown cells ( Figure 4C ) . Furthermore , by only probing the surface M2 protein without permeabilizing the cells , we found that there was no difference in surface M2 protein expression level either ( Figure 4D ) , indicating that CD81 knockdown also did not affect the trafficking of M2 to the cell surface . Similarly , the expression and trafficking of another viral membrane protein , NA , were also not affected upon CD81 depletion ( Figure 4E ) . Taken together , these data indicate that CD81 does not play a direct role in the expression of influenza viral proteins or the trafficking of influenza membrane proteins to the plasma membrane . Next , we probed the role of CD81 in virus assembly . To test whether CD81 is present at the viral assembly sites , A549 cells were infected with the three influenza strains and immunostained for CD81 and the viral protein . Notably , with X-31 infection , CD81 was mostly localized to a site concentrated with multiple viral proteins ( Figures 5A and S4A , B ) . In contrast to uninfected cells , which showed a uniform distribution of CD81 on the plasma membrane , X-31-infected cells exhibited marked redistribution of CD81 into concentrated patches . All of the X-31 proteins that we could obtain specific immunofluorescence staining for , including PB1 , NA and M2 , were present in these patches . The CD81 patches were formed on the plasma membrane , as confirmed by immunofluorescence of non-permeabilized cells ( Figure S4B ) . We note that there was only a modest decrease of CD81 expression upon viral infection ( Figure S5A , B ) . For Udorn-infected cells , CD81 was enriched along the budding virus filaments marked by PB1 ( Figure 5B ) . PB1 is a good filament marker that colocalized with Udorn HA and M2 in the budding filamentous virions ( Figure S4C , D ) . We have also directly observed colocalization between CD81 and other Udorn proteins including HA , NA and an anti-Udorn serum ( Figure S4E–G ) . Remarkably , upon siRNA treatment , which depleted 80∼85% of the endogenous CD81 , the remaining CD81 was all concentrated in the budding viral filaments , whereas the cell body had little CD81 signal ( Figures 5C and S4H ) . The average amount of CD81 per virus filament in the CD81-knockdown cells was reduced by more than 60% compared to that in control cells . Although our lack of WSN antibodies made it difficult to perform similar immunofluorescence experiments on WSN-infected cells , our EM images with CD81 labeled by immunogold showed that CD81 was also recruited to the WSN virus budding zones ( Figure S6A ) . Taken together , these results indicate that CD81 is specifically recruited to the influenza virus assembly and budding sites . To probe which viral component may be responsible for recruiting CD81 , we turned to a plasmid-based system that expresses only specific viral envelope proteins in cells [44] . We transiently transfected the plasmid containing HA or NA in A549 cells and immunostained the cells with CD81 and HA or NA . Interestingly , HA tends to form clusters on the cell surface even when expressed alone in A549 cells and CD81 accumulated substantially in the HA clusters ( Figure S7A ) . About 46% of HA clusters colocalized with CD81 . In contrast , NA when expressed alone did not form clusters but was distributed largely uniformly across the plasma membrane and there was no appreciable correlation between the CD81 distribution and NA distribution ( Figure S7C ) . Mock transfection with plasmid that did not contain HA or NA did not yield any appreciable HA or NA staining ( Figure S7B , D ) . These results suggest that HA is likely responsible for recruiting CD81 to the viral budding sites . Although we observed a ∼50% or more decrease in virus titer in CD81-knockdown cells after acid-bypass treatment to overcome the CD81-dependent entry defects ( Figure 1F ) , it remained unknown whether the defect in viral titer stems from a decrease in the number of budding virions assembled on the cell , the number of progeny virus particles released from the cell , or the specific infectivity per released virus particle . To distinguish between these possibilities , we first infected cells using the acid-bypass treatment and then quantified the number of budding virions attached to the cells using transmission electron microscopy . After quantifying more than 250 cell cross-sections per condition , we performed statistical analysis and found no statistically significant difference in the number of assembling virus particles per cell cross-section in the control versus CD81-knockdown cells ( Figure 6A ) . Next , we infected cells with WSN using the acid-bypass treatment , collected the virus particles in the supernatant , and then quantified the amount of viral matrix protein M1 using an ELISA assay and the number of released virus particles positive of both M1 and HA using an imaging assay ( Figure 6B ) . Notably , compared to control cells treated by non-targeting siRNA , CD81-knockdown cells exhibited 50% or more decrease in both the amount of viral M1 and the number of M1+ and HA+ virus particles released into the supernatant . These results suggest that the CD81-knockdown-induced reduction in viral titer in cells infected by the acid-bypass treatment stems from a defect in virus release . Given that the reduction in viral titer ( Figure 1F ) was similar to the reduction in the amount of released viral proteins or viral particles ( Figure 6B ) , we did not further probe the change in specific infectivity per virus particle . To examine how CD81 may facilitate release of progeny virus particles , we next probed the distribution of CD81 within individual budding virions using immunogold electron microscopy . In X-31-infected cells , CD81 was readily observed in budding virions ( Figure 6C , D ) . During early assembly stages , CD81 clusters located at the growing tip of budding virions ( Figure 6C ) . Interestingly , when viruses grew into a mature , slightly elongated shape [33] , CD81 was not only found on the growing tip , but also on the neck of budding virions ( Figure 6D ) . The elongated morphology of X-31 allowed us to quantitatively analyze the CD81 distribution in virions by aligning the long axis of the virus particles and normalizing the position of immunogold-labeled CD81 to the total length of the virus . Remarkably , CD81 is highly enriched at the two ends of the budding virions ( Figure 6D , E ) . Similarly , we also found CD81 to be enriched in budding WSN virus particles , but the quantity of immunogold detected per WSN virus is substantially lower than that in X-31 viruses , which made it difficult to determine the CD81 distribution in these viruses ( Figure S6A ) . However , the nearly perfectly spherical shape of the WSN virus allowed us to detect an interesting morphological defect of budding viruses in CD81-knockdown cells . When we examined budding virions in cells infected by WSN , we found that most budding WSN viruses were spherical and completely enclosed by viral envelope in control cells treated by non-targeting siRNA ( Figure 6F ) . In stark contrast , budding WSN virus in CD81-knockdown cells appeared much more elongated ( Figure 6G ) . The average length of budding virions in control and CD81-knockdown cells was ∼100 nm and ∼150 nm , respectively ( Figure 6H ) . Furthermore , many budding viruses in CD81-knockdown cells did not have a fully enclosed envelope but remain attached to the plasma membrane through an open membrane neck ( indicated by arrowheads in Figure 6G ) . We performed similar experiments with the X-31 strain . Again , we consistently observed many budding X-31 virions with the open budding neck defect upon CD81 depletion ( Figure S6B , C ) , though characterizing whether the budding virions were further elongated was difficult due to the large variation of the virion length of the pleomorphic X-31 . Taken together , the specific enrichment of CD81 at the neck of the budding virions , the defect in budding neck closure in CD81-knockdown cells , and the reduction in the number of released virus particles but not in the number of assembling virions upon CD81 knockdown suggests that CD81 plays a role in a late stage of the virus budding process , likely at the final scission step . CD81 may facilitate viral scission by directly participating in the scission process , by recruiting host or viral scission proteins , or by organizing the lipid domains and making it conducive to viral scission . Unlike WSN and X-31 strain , Udorn virus infection typically produces filaments that can reach 2 to 20 µm long [11] , [35] , [45] , [46] . We visualized immunogold-labeled CD81 distribution in A549 cells infected by Udorn virus with electron microscopy and observed CD81 clusters in budding viral filaments . Notably , CD81 appeared to be distributed along the entire Udorn filament ( Figure 7A ) . As an alternative approach , we used a super-resolution fluorescence imaging technique , Stochastic Optical Reconstruction Microscopy ( STORM ) , to measure the distribution of CD81 on the budding filament . STORM overcomes the diffraction limit of light microscopy by sequentially activating , imaging and localizing individual fluorescent photoswitchable molecules at high precision , thereby reconstituting images from the molecular localizations with nano-meter scale resolution [47] , [48] . Here , we used a single-objective detection geometry and photoswitchable Alexa Fluor 647 and Atto 488 dyes to obtain a lateral resolution of 20–30 nm and axial resolution of 50–100 nm [48] , [49] . To visualize the localization of CD81 in filamentous Udorn , we immunostained CD81 and viral PB1 , and performed two-color 3D STORM imaging . Consistent with the results from electron microscopy ( Figure 7A ) , we found that CD81 formed small clusters evenly distributed along the entire filament ( Figure 7B ) . Adjacent CD81 clusters were usually separated by about 150∼200 nm . Curiously , CD81 appeared to be enriched between clusters of viral PB1 proteins ( Figure 7B ) , but the significance of this alternating pattern is unclear .
The specific role of CD81 during influenza viral entry was determined using a series of independent assays . First , knocking down CD81 by siRNA led to ∼50% decrease in the percent of infected cells expressing viral proteins . The defect was not due to direct regulation of viral protein expression by CD81 , since viral protein expression remained unchanged upon CD81 knockdown when influenza infection was induced by the acid-bypass treatment ( Figure 1 and Figure 4 ) . These results suggest that CD81 mediates influenza virus entry prior to viral gene expression . Next , CD81 knockdown did not affect virus binding , internalization or trafficking to early endosomes ( Figure 2 ) , but led to a significant defect in viral fusion ( Figure 3 ) . Furthermore , single-virus tracking experiments showed that half of internalized virus particles were trafficked into CD81-positive endosomes and underwent viral fusion within these endosomes , whereas the remaining half fused in CD81-negative endosomes ( Figure 3 ) . Notably , the fraction of viral fusion events occurring within CD81-positive endosomes correlated well with the 50% reduction in the percent of infected cells expressing viral proteins upon depletion of CD81 , suggesting a role of CD81 in productive viral uncoating . Altogether , these results indicate that CD81 plays a role in influenza viral fusion . CD81 marks an endosomal route for productive virus uncoating process , though a parallel CD81-independent route also exists . Interestingly , the role of CD81 in influenza virus entry appears to differ from the role of CD81 previously observed in HCV and HIV entry . As an essential co-receptor for HCV , CD81 is important for the endocytosis of HCV [27] , [28] , [50] , [51] . Furthermore , CD81 interacts with HCV glycoprotein E2 and helps prime its fusogenic activity for low-pH dependent viral fusion [26] , [30] . Moreover , CD81 negatively regulates HIV-cell fusion [52] . Incorporation of CD81 and CD81-associated tetraspanins suppresses the HIV-mediated cellular fusion processes [52] , [53] . On the other hand , CD81 unlikely functions as a co-receptor or attachment factor for influenza viruses because the internalization of influenza viruses into cells does not require CD81 . Influenza viral fusion does not need to be primed by CD81 either , as acid treatment is sufficient to trigger viral uncoating at the plasma membrane in CD81-depleted cells . Instead , our results suggest a role of CD81 in facilitating influenza virus fusion in endosomal compartments . Given that CD81 and CD81-associating proteins can organize membrane domains [17]–[20] , CD81 may help organize the endosomal membrane for assisting influenza viral fusion . Alternatively , CD81 may play a role in trafficking influenza to fusion competent endosomal compartments . CD81 is highly enriched in multivesicular bodies ( MVBs ) , an intermediate endosomal organelle on the maturation pathway of late endosomes and lysosomes [54] . CD81 depletion may inhibit the maturation of endosomes and thus compromise influenza virus fusion with endosomes . In addition to its role in virus uncoating , CD81 also plays a functional role in a later stage of influenza infection post viral fusion . The requirement of CD81 in a post-fusion stage was evident from the finding that CD81 depletion led to a significant decrease in virus titer even when the acid-bypass treatment was used to induce viral uncoating at the plasma membrane , thereby eliminating entry defects ( Figure 1F ) . The decrease did not result from a defect in expression of viral proteins or trafficking of viral proteins to the plasma membrane ( Figure 4 ) , suggesting that the perturbation likely occurred at the virus assembly stage . Furthermore , the average number of budding virions attached to each infected cell did not change upon CD81 knockdown , whereas the number of virus particles released into the supernatant markedly decreased ( Figure 6A , B ) . These results further narrowed the involvement of CD81 to a relatively late stage of the budding process , likely the scission step that severs the virus particle from the host cells . Supporting this notion , CD81 was specifically recruited to viral budding sites ( Figure 5 ) , and among the viral proteins , HA is likely important for recruiting CD81 to the virus budding sites ( Figure S7 ) . Interestingly , CD81 was specifically enriched at the tip and budding neck of the spherical and slightly elongated viruses ( Figure 6C–E ) . Upon CD81 knockdown , the budding spherical viruses exhibited a consistent change in morphology: the budding virions appeared substantially elongated compared to their counterparts in control cells and failed to detach from the plasma membrane . Many budding viruses did not have their budding neck closed , indicating a defect in the final scission process ( Figure 6F–H ) . Taken together , our observations indicate a role of CD81 in scission process that severs the budding virions from the plasma membrane . CD81 could be directly participating in the scission process , recruiting other host or viral scission proteins for this purpose , or organizing the membrane domain at the budding site and making it conducive for viral scission . It is interesting to compare the role of CD81 in the assembly of influenza virus with that of other viruses . Previous studies have shown that HIV envelope proteins associate with a few tetraspanins , including CD81 , and that HIV buds from the tetraspanin-enriched microdomains [55]–[57] . However , the exact role of CD81 in HIV egress remains unclear [53] , [58]–[61] . One study reports that HIV infection is significantly impaired upon CD81 depletion or treatment with anti-CD81 antibodies [61] , whereas two other papers report that depletion of tetraspanins does not affect the efficiency of HIV release whereas overexpression of tetraspanins results in decreased infectivity in released virions [53] , [59] . Tetraspanins have also been proposed to facilitate cell-to-cell transmission of HTLV-1 infection [25] . The role of CD81 in the egress of influenza virus appears different from these previously reported roles of tetraspanins in HIV and HTLV-1 infection in that CD81 positively regulate viral scission . It has been previously shown that influenza virus scission is dependent on viral M2 protein [62] . During virus budding , M2 is localized at the neck of budding viruses and mutation of its amphiphilic tail at the C-terminus leads to a marked defect in virus budding [62] . Interestingly , M2 is known to localize at the interface between lipid rafts and non-rafts region , while CD81 is partitioned into tetraspanin-enriched microdomains , a platform that resembles lipid rafts [18] . Thus , it is possible that CD81 facilitates the recruitment of M2 to the budding neck of the viruses . Future studies on the interaction between CD81 and M2 during the viral scission process would be of interest to further elucidate the mechanistic role of CD81 . CD81 associates with tetraspanins and other tetraspanin-interacting proteins to form tetraspanin-enriched microdomains on cellular membranes [18] , [20] . CD9 , a tetraspanin that interacts with CD81 , was previously identified in the purified virus particles [32] . Our preliminary results revealed that other tetraspanin family proteins were also incorporated into budding viruses ( data not shown ) . Whether and how different components within the tetraspanin-enriched microdomains cooperate with each other in facilitating influenza virus budding remains an interesting question for future investigations .
A549 lung carcinoma cells ( ATCC ) were cultured in high glucose Dulbecco's modified Eagle medium ( DMEM; Invitrogen ) containing 10% fetal bovine serum ( Serum International ) , and antibiotics ( ATCC; 25 U/ml penicillin and 25 µg/ml streptomycin ) , and maintained in humidified , 5% CO2 environment at 37°C . For siRNA knockdown experiments , A549 cells were electroporated with 100 pmol siRNA constructs using program X-001 of Amaxa Lonza Nucleofector with Kit-T ( Lonza , VVCA-1002 ) . Experiments were performed 48 hours post electroporation . siGENOME non-targeting siRNA #1 ( Thermo Scientific ) was used as a control siRNA . Six CD81 siRNA constructs were designed with the following sequences: CD81 siRNA 1: CACCU UCUAU GUAGG CAUCU A dTdT ( Thermo Scientific ) ; CD81 siRNA 2: AAGGA ACAUC AGGCA UGCUA A dTdT ( Thermo Scientific ) ; CD81 siRNA 3: GGAAC AUCAG GCAUG CUAATT ( Qiagen ) ; CD81 siRNA 4: CCUUC UAUGU AGGCA UCUATT ( Qiagen ) ; CD81 siRNA 5: GCCCA ACACC UUCUA UGUATT ( Ambion ) ; CD81 siRNA 6: CCACC UCAGU GCUCA AGAATT ( Ambion ) . Note that CD81 siRNA 1 and CD81 siRNA 2 have been confirmed previously not to cause interferon-induced response [16] . For plasmid expression , 2 µg plasmids were electroporated with a similar procedure . The following plasmids were used in this study: CD81-mEmerald ( human tetraspanin CD81 was cloned into the C terminal of mEmerald , with a 10 amino acid linker between mEmerald and CD81 ) , RFP-Rab5 ( gift from Professor Ari Helenius , Addgene , 14437 [63] ) , EYFP-Rab7 [41] ) , pCAGGS-HA/Ud and pCAGGS-NA/Ud ( gifts from Professor Michael Farzan , Scripps Institute , FL ) . The following viruses were used in this study: influenza virus X-31 was purchased from Charles River Laboratories; WSN and Udorn virus strains were gifts from Professor Robert Lamb ( Northwestern University , Evanston , IL ) . Respiratory syncytial virus was purchased from Virapur . Pseudo-typed MLV virus was a gift from Professor Nir Hacohen ( Broad Institute , Cambridge , MA ) . The following primary antibodies were used in this study: mouse anti-CD81 antibody ( BD Biosciences , 555675 ) , FITC-conjugated anti-CD81 antibody ( BD Bioscience , 551108 ) , mouse anti-EEA1 ( BD Biosciences , 610457 ) , rabbit anti-EEA1 ( Cell signaling , 3288s ) , rabbit anti-CD82 ( Santa Cruz , c-16 , SC-1087 ) , mouse anti-CD63 ( Abcam , ab8219 ) , rabbit anti-EWIF ( Fitzgerald , 70R-13159 ) , mouse anti-ITGB1 ( Millipore , MAB2253 ) , mouse anti-tubulin ( Sigma , T5076 ) , mouse anti-EGFR ( BD bioscience , 610016 ) , rabbit anti-actin ( Abcam , ab8227 ) , rabbit anti-CD9 ( Santa Cruz , H-110 , sc-9148 ) , goat anti-Udorn serum ( gift from Professor Robert Lamb ( Northwestern University , Evanston , IL ) ) , mouse anti-M1 antibody ( AbD Serotec , MCA401 ) , goat anti-M1 antibody ( Abcam , ab20910 ) , mouse anti-influenza virus M2 antibody [14C2] ( Abcam , ab5416 ) , mouse anti-influenza virus NP antibody [AA5H] ( Abcam , ab20343 ) , mouse anti-Alexa Fluor 647 ( Sigma , C1117 ) , mouse anti-RSV fusion protein ( AbD serotec , MCA490 ) , goat anti-influenza virus PB1 antibody ( Santa Cruz , vK-20 ) , mouse anti-influenza virus HA antibody ( Lifespan , LS-C58889 ) , rabbit anti-influenza virus NA ( gift from Professor Gillian Air ( University of Oklahoma , Oklahoma , OK ) ) . The following secondary antibodies were used for immunofluorescence with conventional light microscopy or immunogold electron microscopy: Alexa Fluor 647 donkey anti-mouse ( Jackson ImmunoResearch , 715-605-150 ) , Cy3 donkey anti-mouse ( Jackson ImmunoResearch , 715-165-150 ) , Alexa Fluor 488 donkey anti-mouse ( Jackson ImmunoResearch , 715-545-150 ) , Alexa Fluor 488 donkey anti-rabbit ( Jackson ImmunoResearch , 711-545-152 ) , Cy3 donkey anti-rabbit ( Jackson ImmunoResearch , 711-165-152 ) , Alexa Fluor 647 bovine anti-goat ( Jackson ImmunoResearch , 805-605-180 ) , Alexa Fluor 488 bovine anti-goat ( Jackson ImmunoResearch , 805-545-180 ) , 6 nm gold conjugated goat anti-mouse IgG ( Jackson ImmunoResearch , 115-195-146 ) . The following secondary antibodies were used for immunofluoresence with STORM: donkey anti-mouse ( Jackson ImmunoResearch , 715-005-150 ) labeled with Atto 488 or Alexa Fluor 405 and Alexa Fluor 647 , Bovine anti-goat ( Jackson ImmunoResearch , 805-005-180 ) labeled with Atto 488 or Alexa Fluor 405 and Alexa Fluor 647 . To label antibodies with Alexa Fluor 405 and Alexa Fluor 647 , 80 µl antibody ( 1 . 3 mg/ml ) were mixed with 10 µl 1M NaHCO3 , 8 µg Alexa Fluor 405 and 1 . 2 µg Alexa Fluor 647 dissolved in DMSO for 30 minutes . To label antibodies with Atto 488 , the conditions were similar except 1 . 6 µg Atto 488 was used for the labeling reaction . The mixture was then filtered through a NAP-5 gel filtration column ( GE Healthcare ) to collect labeled antibody . Atto 488 emitted about 1100∼1300 photons per switching cycle , which is significantly lower than that of Alexa Fluor 647 ( 4000∼5000 photons per switching cycle ) [49] . For virus infection , A549 cells were first inoculated with different doses of viruses diluted in DMEM ( without serum ) for 90 minutes at 37°C . Cells were washed with PBS twice to remove unbound viruses , and subsequently incubated with pre-warmed full DMEM medium and maintained at 37°C . For measuring the influenza virus titer ( X-31 , WSN , and Udorn ) without using the acid-bypass treatment , a total of 36 hours were allowed for virus infection before collecting the supernatant for the plaque assay as described below . For RSV virus infection , A549 cells were infected with RSV for 24 hours , followed by immunostaining with anti-fusion protein ( F protein ) antibodies and analysis by flow cytometry . For pseudo-typed MLV virus infection , a 24-hour was allowed for infection and cells were directly fixed to assay the GFP fluorescent intensity by flow cytometry . To ensure equal amounts of viral entry in control and CD81 siRNA treated cells for post-entry studies , an acid-bypass treatment was conducted to induce viral fusion at the plasma membrane . siRNA treated A549 cells were allowed to bind with influenza virus at 4°C for one hour . After extensive washes with cold PBS , a pre-warmed low pH buffer ( PBS , pH 4 . 5 ) was added in for two minutes . The low pH buffer was then neutralized with culture medium and cells were placed with pre-warmed fresh culture medium afterwards . For measuring the influenza virus titer ( X-31 and WSN ) with the acid-bypass treatment , a total of 17 hours were allowed for virus infection before collecting the supernatant for the plaque assay . Cells were infected with influenza virus for indicated amounts of time as described above and supernatant was collected to assay the virus titer at the end of each time point . Serial dilutions of the supernatant were used to inoculate MDCK-Texas cells ( Kind gift from Robert Lamb ) seeded in 6-well plates for 90 minutes at 37°C . After washing with PBS twice , a 3 ml agar overlay of DMEM containing 30% Noble agar ( Affymetrix ) , antibiotics and 2 µg/ml acetylated trypsin ( Sigma ) was placed on cells . The plates were incubated at 37°C . After about two days , the agar disks were removed carefully and cells were immediately stained with crystal violet solution ( 1∶1 , 000 V/W crystal violet , 30% ethanol in water ) for 10 to 15 minutes , which allowed for an easy quantification of the number of plaques . Virus titer ( PFU/mL ) = number of plaques/ ( dilution factor×inoculation volume ( mL ) ) . For each condition , samples were tested with triplicates . Influenza virus X-31 was either labeled with lipophilic dye DiD ( Invitrogen , D7757 ) or Alexa Fluor 647 ( Invitrogen , A-20006 ) as previously described [42] . For the labeling reaction , 100 µl of the original virus stock ( 2 mg/ml protein concentration ) was incubated with either 3 µl of 25 mM DiD or 3 µg Alexa Fluor 647 dissolved in DMSO for two hours or one hour respectively with gentle vortexing in the dark at room temperature . Unincorporated dye was removed by buffer exchange into the Hepes 145 buffer ( 50 mM Hepes , pH 7 . 4 , 145 mM NaCl ) by using NAP-5 gel filtration columns ( GE Healthcare ) . The labeled virus was aliquoted , snap-frozen in liquid nitrogen , and stored at −80°C . Immediately before experiments , the labeled virus was thawed and filtered through a 0 . 2 µm pore size syringe filter ( Supor membrane , Pall ) to remove viral aggregates . Labeled viruses are infectious , as confirmed with standard plaque assays ( data not shown ) . Control or CD81-knockdown A549 cells were allowed to bind with DiD-labeled X-31 diluted in DMEM ( without serum ) for 30 minutes at 4°C . After extensive washes with cold PBS to remove unbound viruses , cells were trypsinized and immediately fixed with 2% paraformaldehyde ( PFA ) for 20 minutes at room temperature . After washing PFA away with PBS , the DiD fluorescent intensity was measured by a flow cytometer ( BD bioscience ) . At least 10 , 000 cells were quantified for each measurement . The data was analyzed via FlowJo . Control or CD81-knockdown A549 cells were allowed to bind with 3×104 PFU/ml Alexa Fluor 647-labeled X-31 virus diluted in DMEM ( without serum ) on ice for 30 minutes at 4°C . After extensive washes with cold PBS to remove unbound viruses , pre-warmed full culture medium was added in and the virus was allowed to internalize at 37°C for indicated amounts of time . At the end of each time point , cells were washed with PBS , directly fixed with 4% PFA for 20 minutes at room temperature . In order to distinguish the surface-bound versus internalized virus particles , a non-permeablizing immunofluorescence in the absence of detergents was performed by using a mouse anti-Alexa Fluor 647 primary antibody ( Sigma , C1117 ) , followed by staining with an Alexa Fluor 555-conjugated donkey anti-mouse secondary antibody ( Invitrogen , A31570 ) . The samples were imaged using a custom-built spinning disk confocal microscope . Non-internalized virus particles were stained with both Alexa Fluor 647 and Alexa Fluor 555 , while the internalized virus particles—inaccessible to the antibodies—exhibited only the Alexa Fluor 647 signal . To quantify the number of particles internalized per cell , we used the maximum z-projection of the confocal z-stacks , and counted both the total number of virus particles ( with Alexa Fluor647 signal ) and the number of non-internalized virus particles ( with both Alexa Fluor 647 and Alexa Fluor555 signals ) , and the number of internalized particles ( with Alexa Fluor 647 but not Alexa Fluor 555 signal ) . A low enough number of virus particles ( ∼15 particles ) was internalized each cell to minimize the possibility of multiple virus particles sorting into the same vesicle . Statistical analysis was performed using a two-tailed student t-test . Cell lysate samples were prepared with Laemmli sample buffer ( Bio-Rad , 161-037 ) and run on a 4–15% Tris-HCL polyacrylamide gel ( Bio-Rad ) . After transferring the protein onto Hybond polyvinylidene difluoride membranes ( GE Healthcare ) , the membrane was blocked with 5% nonfat-milk in TBS-Tween for 1 hour , followed with incubation of primary antibody overnight at 4°C , a three 10 minutes wash step with TBS-Tween , and a one hour incubation of HRP-conjugated secondary antibody at room temperature . The signal was detected with TMA-6 ( TMA-100 , Lumigen ) and developed to Kodak films . Note that CD81 and CD63 could only be detected under non-reducing conditions . For flow cytometry analysis , the procedures were similar to what was previously described [64] . Briefly , for measuring total protein expression level ( CD81 or viral proteins ) , cells were collected and fixed with 2% PFA for 20 minutes . After washing with PBS once , cells were permeablized with buffer P ( 0 . 075% Saponin , 10% BSA in PBS ) for 20 minutes at room temperature . Cells were incubated with primary antibodies diluted in buffer P ( 1∶1 , 000 ) for 1 hour and washed with buffer P three times before incubating with secondary antibodies for another 45 minutes . Secondary antibodies were also diluted with buffer P ( 1∶1 , 000 ) . After washing with buffer P , cells were resuspended with PBS and then analyzed by flow cytometry . The data was analyzed by FlowJo . Cells were gated based on FSC and SSC scattering , and a histogram was generated based on the fluorescence intensity profile . For cells that were infected with influenza virus , a second gate was set based on comparison of fluorescence intensity of uninfected versus infected cells . The population that falls into the second gate corresponds to the percent of infected cells in each sample , from which the mean fluorescence intensity was analyzed to infer the viral protein expression level . To probe surface protein expression , all steps were similar except detergent was excluded . For imaging-based experiments , A549 cells seeded in Lab-Tek 8 well glass dishes were fixed with 4% PFA for 20 minutes at room temperature . Unless specified , fixed cells were permeablized with 0 . 1% Triton-X100 in PBS for 5 minutes , washed with PBS twice and incubated with blocking buffer PBSA ( 3% BSA in PBS , or 5% bovine serum in PBS ) for 30 minutes . Cells were then incubated with primary antibodies diluted in PBSA ( 1∶500 ) for 1 hour . Followed by three PBS washes ( 5 minutes each ) , secondary antibodies were added for another 1 hour . Afterwards , cells were washed with PBS for three times again before imaging . For STORM , a post-fixation step was followed with 3% PFA and 0 . 1% glutaraldehyde ( GA ) in PBS for 20 minutes . For immunostaining surface protein only , permeablization was not performed after fixation . When antibody species conflict existed , labeled primary antibodies ( CD81-FITC , BD Bioscience; HA-Alexa Fluor 647 ) were used as needed . To quantify the colocalization ratio between internalized virus and cellular proteins ( EEA1 and CD81 ) , samples were prepared similarly as in virus internalization assay . After probing the surface-bound virus particles with anti-Alexa Fluor 647 antibody , cells were permeabilized with 0 . 1% Triton-X100 in PBS , and a subsequent indirect immunofluorescence was performed to stain against each protein . Images were acquired by confocal microscopy and at least 40∼100 randomly chosen cells were analyzed manually for each condition . Only internalized virus particles were used to quantify the fraction of viruses colocalized with CD81 or EEA1 . Cells were infected with influenza virus with acid bypass for 17 hours and the supernatant was collected to assay the total amount of viral protein in the released virus particles . M1 was chosen due to its abundance in the virus to maximize signal . Nunc 96 well plates ( eBoscience , 44-2404-21 ) were incubated with capture antibody ( Goat anti-M1 , Abcam , 1∶1000 diluted in 0 . 2 M sodium carbonate/bicarbonate buffer , pH 9 . 4 ) at room temperature for 2 hours . After three 5 minutes wash with PBST ( 0 . 05% Tween in PBS ) , the plates were blocked with PBSA ( 2% BSA in PBST ) for 1 hour . The supernatant was mixed with RIPA buffer ( 1∶2 dilution ) , and added in each well for overnight incubation at 4°C . The samples were washed three times with PBST , and incubated with detection antibody ( Mouse anti-M1 , AbD Serotec , 1∶600 dilution in PBSA ) for 1 hour at room temperature , washed three times , and further incubated with HRP-conjugated goat anti-Mouse antibody ( Bio-Rad , 172-1011 , 1∶5 , 000 ) for 1 hour . TMB substrate ( Thermo Scientific , N301 ) was used to detect HRP activity , and the reaction was stopped by 0 . 18 M sulfuric acid before measuring the absorbance at 450 nm . The experiment was performed with triplicate samples from independent infections , with three measurements for each sample . To confirm the efficiency of detection , purified X-31 virus ( Charles River laboratory ) was used as a standard sample ( data not shown ) . Control and CD81 siRNA treated cells were infected with influenza virus by acid bypass treatment . At 17 hours post infection , supernatant was collected to assay for the number of released virions through immunofluorescence . Briefly , the supernatant was absorbed on poly-lysine coated Lab-Tek 8 well glass dishes at 4°C overnight . After washing away unbound virions with PBS twice , the sample was fixed with 4% PFA for 15 minutes , blocked with 3% PBSA buffer , immunostained with anti-HA and anti-M1 antibody , and then imaged by confocal microscopy . More than 150 randomly selected regions were imaged , and the number of particles positive for HA and M1 staining was quantified ( with more than 600 virus particles ) . The single-particle tracking experiment has been described in detail previously [5] , [41]–[43] . Briefly , A549 cells were nucleofected with CD81-mEmerald or RFP-Rab5 plasmids 24 hours prior to single virus tracking experiments . After washing the cells with pre-warmed PBS twice , 2 . 6×104 PFU/ml DiD-labeled X-31 virus diluted in imaging buffer ( 9 parts DMEM without phenol red , 1 part pH 8 Hepes buffer , supplemented with oxygen scavenge system: 0 . 8 mg/ml dihydroxybenzoic acid ( PCA , Sigma , 37580 ) , 0 . 5 U/ml protocatechuate 3 , 4- dioxygenase ( PCD , Sigma , P8279 ) ) was added to the cells . The objective and stage of the microscope were heated to maintain the temperature at 37°C for the cells . Image acquisition began immediately after adding DiD-labeled virus in situ . To obtain a simultaneous imaging of DiD-labeled virus and cellular protein , DiD was excited with a 647 nm krypton laser ( Coherent ) while mEmerald and RFP was excited with a 488 nm argon ion laser ( Coherent ) and a 561 nm solid state laser ( CrystaLaser ) respectively . Fluorescence emissions from DiD and mEmerald/RFP were separated by a 630 nm long-pass dichroic , filtered with bandpass filters ( 705/40 for 647 channel , 525/40 for 488 channel , and 605/70 for 561 channel ) and imaged on a EMCCD camera ( Andor ) with 0 . 5 second exposure time . The imaging analysis was performed as described previously [5] . Briefly , for each image , the fluorescence signal collected from the DiD channel was convolved with a Gaussian spatial filter to remove background and noise . To identify the virus peaks , the algorithm performs recursive integration over bright regions connected to each local maximum . The centroid of each fluorescent peak was computed to determine the virus particle position , and the trajectories were obtained by reconstructing paired peaks between adjacent frames with similar proximity and intensity . The fluorescence intensity of DiD was plotted versus time . Control or CD81-knockdown A549 cells seeded in Lab-Tek 8 well glass dishes were rinsed with cold PBS first , and then incubated with 2×104 PFU/ml DiD-labeled X-31 diluted in DMEM ( without serum ) for 45 minutes at 4°C . After washing away unbound viruses with cold PBS twice , pre-warmed full culture medium was added in and cells were maintained at 37°C for indicated amounts of time . At the end of each time point , cells were trypsinized and immediately fixed with 2% PFA for 20 minutes . Afterwards , cells were washed and resuspended with PBS . DiD-fluorescent intensity was measured through a flow cytometer . At least 10 , 000 cells were measured for each measurement with duplicates for each condition . The data was analyzed with FlowJo . The normalized viral fusion extent ( mean DiD intensity at each time point-initial mean DiD intensity ) /Initial mean DiD intensity ) was plotted . STORM experiments were performed as previously described on an Olympus IX71 inverted optical microscope [65] . Three lasers were used in this study for STORM: 657 nm ( RCL-300-656; Crystalaser ) , 488 nm ( Sapphire 460-10; Coherent ) and 405 nm ( CUBE 405-50C; Coherent ) . A high-numerical-aperture ( NA ) oil-immersion objective ( 100× UPlanSapo , NA1 . 4; Olympus ) was used to collect the fluorescence emission , which is imaged onto a back-illuminated electron-multiplying charge-coupled device ( EMCCD ) camera ( iXON DU-897; Andor ) . For two-color 3D imaging of Alexa Fluor 647 and Atto 488 , two imaging laser beams ( 488 nm and 657 nm ) and an activation laser beam ( 405 nm ) were reflected by a custom-designed polychroic mirror ( z488/647/780rpc; Chroma ) . Fluorescence emission from Alexa Fluor 647 and Atto 488 were separated by a 630 nm long-pass dichroic mounted on a commercial beamsplitting device ( 3D Dual-View with a cylindrical lens , 100 cm focal length; Photometrics ) . Two bandpass filters: FF01-535/50 ( Semrock ) and ET705/72m ( Chroma ) were used to filter for the short-wavelength and long-wavelength channel independently . In addition to the bandpass filters , a double-notch filter ( NF01-488/647; Semrock ) was added before the Dual-View ( Photometrics ) to block the two excitation laser beams . STORM imaging for each channel was performed at 60 Hz sequentially and each channel was imaged onto 256×256 pixels in the EMCCD camera ( iXON DU-897 ) . STORM images were generated using similar methods as previously described [65] . The STORM images in the Alexa Fluor 647 and Atto 488 channels were aligned by a third-order polynomial warping map in three dimensions obtained from calibration images of 100-nm Tetraspeck fluorescent beads . The residual alignment error was ∼7 nm in x-y and ∼20 nm in z dimensions . To correct for the sample drift during imaging acquisition , we relied on the correlation function of imaging itself to correct for the lateral and axial drift , as previously described [48] . Since the imaging acquisition was performed sequentially with the longer wavelength channel first , 647 channel was drift-corrected to the last frame of the 647 nm acquisition while 488 channel was drift-corrected to the first frame of the 488 nm acquisition . The spatial resolution measured was 20∼30 nm laterally and 50∼60 nm axially for Alexa Fluor 647; while Atto 488 gives a lateral resolution of 30∼40 nm and axial resolution of ∼100 nm . A549 cells were infected with influenza viruses at a MOI of 2 . Infection was allowed to proceed for a total of 15 hours before fixing with 2 . 5% PFA/GA in 0 . 1M sodium cacodylate buffer , pH 7 . 4 ( Electron Microscopy Sciences , 15949 ) at room temperature for at least 1 hour . The cells were post-fixed for 30 minutes in 1% Osmium tetroxide ( OsO4 ) /1 . 5% potassium ferrocyanide ( KFeCN6 ) , washed in water three times and incubated in 1% aqueous uranyl acetate for 30 minutes followed by two washes in water and subsequent dehydration in grades of alcohol ( 5 minutes each; 50% , 70% , 95% , 2×100% ) . Cells were removed from the dish in propyleneoxide , pelleted at 3000 rpm for 3 minutes and infiltrated for 2 hours in a 1∶1 mixture of propyleneoxide and TAAB Epon ( Marivac Canada Inc . St . Laurent , Canada ) . The samples subsequently embedded in TAAB Epon and polymerized at 60°C for 48 hours . Ultrathin sections ( about 60 nm ) were cut on a Reichert Ultracut-S microtome , transferred onto copper grids stained with lead citrate and examined in a TecnaiG2 Spirit BioTWIN and images were recorded with an AMT 2k CCD camera . For immunogold electron microscopy , at the end of indicated time points of virus infection , cells were rinsed with PBS once , fixed with 4% PFA for 15 minutes , and blocked with PBSA ( 3% BSA in PBS ) for 30 minutes . Primary antibodies diluted in PBSA ( 1∶500 ) were incubated with cells for overnight at 4°C . After three washes with PBS , cells were treated with 6 nm gold-conjugated secondary antibodies ( 1∶40 ) for four hours , post-fixed with 2 . 5% PFA/GA in 0 . 1 M sodium cacodylate buffer for at least one hour . The sample were embedded and sectioned as described above for transmission electron microscope imaging .
|
As a “Trojan Horse” that only encodes 13 viral proteins , influenza hijacks host cell machinery for productive infection . In this work , we studied the role of the host protein CD81 in influenza infection . We found that CD81 was important for influenza infection at two distinct stages: virus uncoating and virus budding . Specifically , during virus entry , more than half of internalized virus particles were trafficked into a specific CD81-positive endosomal population for virus uncoating . Depleting CD81 led to a significant defect in viral uncoating and infection . During virus egress , CD81 was recruited to virus assembly site , and incorporated into individual virions at specific sub-viral locations . CD81 depletion resulted in virions that failed to detach from the plasma membrane and a marked decrease in progeny virus production .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Dual Function of CD81 in Influenza Virus Uncoating and Budding
|
Although cholera is a major public health concern in Mozambique , its transmission patterns remain unknown . We surveyed the genetic relatedness of 75 Vibrio cholerae isolates from patients at Manhiça District Hospital between 2002–2012 and 3 isolates from river using multilocus variable-number tandem-repeat analysis ( MLVA ) and whole genome sequencing ( WGS ) . MLVA revealed 22 genotypes in two clonal complexes and four unrelated genotypes . WGS revealed i ) the presence of recombination , ii ) 67 isolates descended monophyletically from a single source connected to Wave 3 of the Seventh Pandemic , and iii ) four clinical isolates lacking the cholera toxin gene . This Wave 3 strain persisted for at least eight years in either an environmental reservoir or circulating within the human population . Our data raises important questions related to where these isolates persist and how identical isolates can be collected years apart despite our understanding of high change rate of MLVA loci and the V . cholerae molecular clock .
Cholera remains a public health concern in developing countries with an estimated burden of 1 . 2–4 . 3 million cases and 28 , 000–142 , 000 deaths per year , worldwide [1] . South Asia and sub-Saharan Africa account with the majority of cases and deaths . Between 01 January and 03 June 2013 , a total of 25 , 762 cholera cases and 490 deaths were reported from 18 African countries . Mozambique accounted for 7% ( 1 , 861/25 , 762 ) of cases and 4% ( 19/490 ) of deaths , being the third most affected country after the Democratic Republic of the Congo and Angola [2] . The most recent cholera outbreak in Mozambique started in December 2014 in Nampula [3] , where there were 8 , 835 cases with case fatality rate of 0 . 7% ( 65 deaths ) in 5 northern and central provinces in five months [4 , 5] . The most recent cholera outbreak in the south was reported in 2011 [6] . The peak of a cholera epidemic is often preceded by increasing prevalence of the pathogenic strains in the environment [7] where V . cholerae are harbored in aquatic reservoirs during extended periods between outbreaks [8] . Vibrio cholerae O1 is associated with most epidemic and pandemic outbreaks [9] . Whole genome sequence ( WGS ) analysis of V . cholerae isolates from around the world [9–12] demonstrated that the current ( Seventh ) Pandemic is monophyletic and originated from a single source with a clonal expansion of the lineage , with a strong temporal signature [13] . The seventh pandemic has been divided into 3 waves beginning in 1952 , 1981 and 1988 , respectively [13] . Each wave appears to be a selective sweep , as was the second wave of V . cholerae that swept across Haiti after the initial introduction [14] . In Africa , isolates from all three waves have been identified [13] with wave 3 isolates forming two distinct clades in Kenya between 2005 to 2010 [15] . The epidemiology and transmission patterns of outbreaks of V . cholerae have been explored using both multilocus variable number tandem repeat analysis ( MLVA ) and WGS [13 , 16] . In rural Bangladesh , MLVA revealed that multiple genetic lineages of V . cholerae occur naturally in the environment with geographic and seasonal genetic variation and identical genotypes can be found in the environment and humans [12] . In Kenya , MLVA demonstrated that several distinct genetic lineages emerged simultaneously during outbreaks in a single cholera season , linked to local environmental reservoirs [17] . WGS revealed all lineages were part of the Seventh Pandemic expansion [15] . In central and western Africa , MLVA revealed distinct clusters of isolates from different countries: Democratic Republic of the Congo , Zambia , Togo , and Guinea [18] . A second study in Guinea demonstrated spread and differentiation of V . cholerae during an outbreak [19] . In Mozambique , previous analysis of V . cholerae O1 showed these isolates had the typical traits of the El Tor biotype overall except that they carried a tandem array of classical CTX prophage [20] located on the small chromosome [21 , 22] . However , there are no data available on the genetic relatedness of V . cholerae circulating in Manhiça in southern Mozambique . Here , we characterized clinical and environmental V . cholerae isolates from Mozambique using MLVA and WGS to determine the genetic relatedness of strains isolated from patients with diarrhea in Manhiça District Hospital .
Manhiça District Hospital ( MDH ) is a 110 bed referral health facility for Manhiça District , a rural area of Maputo Province in southern Mozambique . The characteristics of the area have been described in detail elsewhere [23 , 24] . Briefly , the climate is subtropical with two distinct seasons: a warm , rainy season between November and April , and a cool and dry season during the rest of the year . Manhiça has 160 , 000 inhabitants , who are mostly subsistence farmers or workers in two large sugar- and fruit-processing factories . The Manhiça Health Research Centre ( Centro de Investigação em Saúde da Manhiça[CISM] ) is adjacent to the MDH and has been conducting continuous demographic surveillance for vital events and migrations since 1996 [23] , currently covering 165 , 000 individuals . The strain collection of V . cholerae described in the study was isolated from cholera surveillance and other studies conducted in the Manhiça community by CISM approved by the National BioEthic Committee ( CNBS ) . Any and all patient data were anonymized/de-identified . The IRB at University of Maryland School of Medicine approved the use of anonymized strains . A total of 75 V . cholerae isolates were collected from MDH , between 2002 and 2012 . The strains were isolated from stool of patients admitted to the MDH with suspicion of cholera , presenting with watery diarrhea . In addition , three isolates collected from the Incomati River were included . V . cholerae isolates were identified by standard biochemical tests; and confirmed by API-20E biochemical test strips ( bioMérieux SA , Marcy-l'Etoile , France ) . Serotypes were determined using commercially available poly- and mono-clonal slide agglutination antisera ( Mast Group Ltd . , Merseyside , UK ) according to the manufacturer’s instructions . All the isolates were stored at -80°C in tryptone soya broth ( TSB ) with 15% glycerol , and retrieved at the time for molecular characterization . A pure culture of V . cholerae was plated in Thiosulfate Citrate Bile Sucrose ( TCBS ) agar and incubated overnight at 37°C . DNA was extracted using the Qiagen QIAamp DNA Mini Kit ( Hilden , Germany ) . The DNA template was sent to the University of Maryland Baltimore , Baltimore , Maryland , USA for molecular typing by MLVA and WGS . DNA from each isolate was amplified by PCR using the conditions and primers previously described for 5 loci containing variable length tandem repeats [11] . The amplified products were separated and detected using a model 3730xl Automatic Sequencer ( ABI ) and their sizes were determined using internal lane standards ( Liz600; ABI , Foster City , CA ) with the Gene Mapper v4 . 0 program ( ABI ) . Genotypes were determined according to the published formulas to calculate the number of repeats from the length of each allele and identify the alleles at the 5 loci [11] . The 5 loci , in order , are VC0147 , VC0436–7 ( intergenic ) , VC1650 , VCA0171 , and VCA0283; thus , the genotype 9 , 4 , 6 , 19 , 11 indicates that the isolate has alleles of 9 , 4 , 6 , 19 , and 11 repeats at the 5 loci , respectively . Relatedness of the strains was assessed by eBURSTv3 ( http://eburst . mlst . net ) , in which genetically related genotypes were defined as those possessing at least 4 identical alleles of the 5 loci . An alternative analysis was performed using Network 2 . x ( http://www . fluxus-engineering . com/sharefaq . htm ) . The DNA concentration was quantified by NanoDrop 2000 Spectrophotometer ( Thermofisher Scientific , Waltham , MA , USA ) and only specimens with sufficient concentration ( n = 71 ) were submitted to WGS . DNA was prepared for Illumina sequencing using the KAPA High Throughput Library Preparation Kit ( KapaBiosystems , Wilmington , MA ) . DNA was fragmented with the Covaris E210 . Libraries were prepared using a modified version of manufacturer’s with-bead protocol ( KapaBiosystems , Wilmington , MA ) . The libraries were enriched and barcoded by ten cycles of PCR amplification step with primers containing an index sequence seven nucleotides in length . The libraries were sequenced on a 100 bp paired-end run on an Illumina HiSeq2500 ( Illumina , San Diego , CA ) . The quality of the 101-base paired-end reads was confirmed using Fastqc ( https://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) . Kmergenie ( http://kmergenie . bx . psu . edu/ ) was used for choice of the best peak and the assembly was performed using the SPAdes software [25] . CSI Phylogeny 1 . 0a ( http://cge . cbs . dtu . dk/services/CSIPhylogeny/ ) was used to generate a tree of genetic relatedness based on high quality nucleotide variants and then compared to V . cholerae O1 El Tor reference strain N16961 ( NCBI accession numbers AE003852 and AE003853 ) . Splitstree [26] was used to determine networks . Previous work in our lab demonstrated that this pipeline produces identical results to SMALT [27] . A high resolution SNP based phylogeny for the 67 7th pandemic strains was placed in context of a globally representative collection of 274 isolates ( S1 Table ) by mapping the reads to the V . cholerae 01 El Tor reference N16961 using SMALT ( http://www . sanger . ac . uk/resources/software/smalt ) as previously described [28] . Gubbins [29] was used to simultaneously remove regions of high SNP density and putative recombination sites in the alignment and infer the phylogenetic tree . The pre-seventh pandemic strain M66 ( NCBI accession numbers CP001233 and CP001234 ) was used to root the phylogenetic tree . To accurately place the four non-01 V . cholerae into a phylogenetic context we calculated a core genome alignment of 1093 genes ( 1 , 055 , 747 bp ) using Roary [30] from a set of diverse V . cholerae genomes along with genomes of the closely related species Vibrio metoecus and Vibrio parilis . The resulting alignment was used to reconstruct the phylogenetic relationship using RAxML v . 7 . 8 . 6 [31] under the GTR model with 100 bootstrap replicates . The resulting phylogenetic trees were visualized using FigTree v . 1 . 4 . 2 ( http://tree . bio . ed . ac . uk/software/figtree/ ) .
Among the 78 isolates ( Table 1 ) , 26 distinct genotypes were identified by MLVA using all five VNTR loci . The numbers of distinct alleles at loci VC0147 , VC0437 , VC1650 , VCA0171 , and VCA0283 were 10 , 2 , 4 , 7 and 7 , respectively . Two clonal complexes of genetically related genotypes ( each complex comprising genotypes that differed by an allelic change at a single locus ) and four singleton genotypes unrelated to any other ( differed by ≥2 loci ) were identified ( Fig 1 ) . The first clonal complex ( CC1 ) consisted of twenty of the twenty-six genotypes and comprised 91% ( 71/78 ) of the isolates ( including all 3 isolates from river water ) . In CC1 the most common genotype , identified as the “founder” genotype ( defined by eBURST as the genotype that differed from the largest number of other genotypes at a single locus ) , comprised 41% ( 29/71 ) of the isolates . In this complex , the founder genotype radiated into 9 other genotypes , and 5 of those differentiated further . Two of the three river water isolates shared a common genotype with the clinical isolates , one of which was of the founder genotype . Interestingly we found isolates with identical MLVA genotypes up to 8 years apart . In general , isolates from the same year were genetically related , all four genotypes found in 2002 were single locus variants , as were the two genotypes in 2008 , two in 2009 , and three genotypes from 2010 . In contrast to similarity within a year , in months ( May 2002 , March 2003 , April 2003 , May 2003 , June 2010 ) where multiple cases presented with V . cholerae , we found that isolates belonged to more than one genotype . The other V . cholerae isolates consisted of a second clonal complex CC2 and four singletons ( Fig 1 ) . CC2 had two genotypes and comprised only 4% ( 3/78 ) of the analyzed isolates . The four singleton genotypes contained one isolate each corresponding to 5% ( 4/78 ) of the isolates . Three of them were isolated in 2008 and the fourth in 2012 . We successful sequenced the whole genomes of 71 V . cholerae O1 isolates ( Table 1 ) . The average number of high quality reads was 15 , 375 , 291 which upon assembly produced an average of 143 contigs ( range: 99 to 443 ) or scaffolds with an average depth of 374 . The assembled genome was 4 . 09 Mb ( 3 . 95 to 4 . 23 Mb ) in length and had an average N50 of 232 , 090 bp ( 144 , 507 to 339 , 296 bp ) . Of the 71 isolates , the genomes of sixty-seven differed by less than 100 SNVs . Consistent with this , the core genome of all the wave 3 isolates shared 2802 genes . To look for evidence of recombination in these sequences we examined 2543 single copy conserved genes . We removed all of the invariant nucleotides , any SNVs in locally collinear blocks smaller than 200 bp , and examined the remaining variant sites in Clustal W ( Fig 2 ) . These data showed evidence of homoplasy by visual examination: for example there were multiple examples of dinucleotide sequences at the same site in different isolate genomes being present in all four possible combinations ( e . g . AA , AG , GA , GG ) . This was despite the high level of pairwise nucleotide conservation between isolate genomes , usually differing by only 18 to 62 nucleotides . Among the 148 SNVs , 360 pairs were determined to have both alleles in all possible combinations . The probability of this occurring by mutation alone is vanishingly small ( the probability of the same mutation on two different genomes to the 360th power ) . The most parsimonious explanation for this observation is recombination within this population of V . cholerae O1s . The genetic relatedness , including the recombinant loci , was estimated from WGS using a phylogenetic network [26] . As shown in Fig 3 , while all of the genomes were distinct , there was some clustering by year of isolation , four of the five isolated in 2010 clustered together as did the four from 2008 and the two from 2009 . The largest number of isolates in our collection were collected in 2003 . It is evident from Fig 3 that they occupy positions throughout the network . This was also true for the three river isolates . In the network , every sequence had some nucleotide variants that distinguish it from every other sequence , thus in contrast to the MLVA network , there are no central genotypes that might be construed as the founder . Fig 4 shows the phylogenetic tree with genomic sequences from a global collection of 274 previously published sequences ( S1 Table ) including the 67 genomic sequences generated in this study . These data show that the sequences of the Manhiça , Mozambique isolates cluster in a monophyletic group distinct by 39 nucleotides from the backbone of the third wave of the Seventh Pandemic phylogeny ( S2 Table ) . All Seventh Pandemic isolates harbor virulence associated genes , such as the CTX prophage , the genomic islands VSP-I and VSP-II , the toxin-coregulated pilus , the toxic linked cryptic element , and the integrative conjugative element SXT harboring multidrug resistance genes . The CTX phage harbours the ctxBcla allele within an otherwise El Tor biotype CTX phage sequence and is typical of isolates referred to as “atypical” El Tor biotypes [32] . The four sequences that differed by >24 , 000 nucleotides among the 3 , 237 , 973 basepairs conserved in all our sequenced genomes did not cluster with sequences from the Seventh Pandemic ( Fig 5 ) and are quite distantly related to each other , but all were more closely related to the other V . cholerae sequences compared to other species . These isolates were taken from patients admitted to hospital with suspected cholera indicating that these divergent lineages are capable of causing clinical symptoms , as has been reported previously [13] . Unlike the Seventh Pandemic strains , these four strains do not contain any of the aforementioned virulence associated genes , although they contain a few genes , such as the RTX toxin gene cluster and the hemolysin hlyA , from some of the virulence associated islands . The network based on WGS included isolates of both CCs ( CC1 and CC2 ) , however while the isolates of CC1 was distributed through the network , the isolates of CC2 clustered together ( Fig 3 ) . In the same way , the four singletons isolates by MLVA did not cluster in the network and were quite distantly related by WGS analysis .
Cholera remains important public health problem in Mozambique . We characterized V . cholerae isolates using MLVA and WGS to determine the genetic relatedness and transmission dynamics of cholera outbreaks in the Manhiça District . Our analyses of WGS data revealed that 94% of the isolates were a monophyletic group in the third wave of the Seventh Pandemic . These isolates have formed a locally evolving population that has persisted for at least eight years , either in a local environmental reservoir or circulating within the human population , and sporadically caused recurrent disease in southern Mozambique . It is yet to be shown if these isolates are representative of those circulating outside of the study site across Mozambique . Previous studies have reported the role of environmental factors , such as seasonal fluctuations , that influence the dynamics of V . cholerae in environmental reservoirs [33 , 34] . Of note , all of our isolates were collected during the first half of the year , January through July , with a peak in May . However , our study does not allow establishing a clear relationship between environmental strains and those causing cholera outbreak . The presence of four isolates that are not part of the Seventh Pandemic ( differing by ~ 0 . 75% of the genome sequence ) demonstrates that a diverse set of V . cholerae not linked to the Seventh Pandemic are causing a background , low level of sporadic disease in southern Mozambique , despite the absence of many of the major virulence determinants . This has been seen elsewhere including in the Gulf coast of the USA particularly in the 1980’s and 1990’s [32] . We detected the presence of recombination in our WGS data . Most analyses of WGS data detect recombination as a large number of SNVs in short sequence of DNA . Our method of detection , the four-gamete test applied to haploid genomes , removes the constraint of the SNVs occurring in a small region by simply looking at dinucleotide pairs located at any distance from each other . Since mutations occur at random , a dinucleotide sequence , say GG , can mutate at either of two positions for example: AG and GA . In order to further mutate to form AA , one of the positions must mutate a second time . However , in this instance it also possible to replace the original sequence in a single recombination event . We found 360 pairs of nucleotide positions across all our genomes at which all four combinations of alleles in the dinucleotides were found . The probability of this occurring by mutation is extremely small ( the rate of mutation at the same nucleotide to the 360th power ) . Our sample is unusual because it was collected in small area and in a short time frame . The geographical and temporal proximity of isolates is a prerequisite for recombination . V . cholerae meets other necessary prerequisites like having an intact mechanism for uptake of DNA and integration into the chromosome [38] . The clearest examples of the effectiveness of recombination can be found in serotype switching , a process known to be accelerated by chitin [39] . The limited amount of variation in our sample is consistent with the recombinant events occurring within this population of V . cholerae O1s . Most isolates ( 91% ) were distributed among the 20 genotypes of CC1 and 41% had the founder genotype , supporting the hypothesis of a common ancestor which subsequently differentiates into additional genotypes . In a study of 187 isolates conducted in Haiti , only 9 MLVA genotypes clustered in a single clonal complex and 53% had the founder genotype [35] . We found more alleles ( ten ) at the first locus ( VC0147 ) on large chromosome than at the two loci on small chromosome ( VCA0171 and VCA0283 ) with seven alleles for each one . Our findings are in contrast to previous descriptions that show the three loci on the large chromosome varying at a slower rate than those on the small chromosome [11] . Extensive genetic variation on both chromosomes has been reported [36] , but not more variation at the large chromosome loci [35–37] . WGS and MLVA patterns were performed to discriminate V . cholerae isolates from various geographic locations and distinct populations [17 , 40] . But , none of the 26 MLVA genotypes that we found in this study has been reported in previous studies from Haiti , Thailand , Bangladesh , India , Vietnam and Mozambique [10 , 11 , 35 , 36 , 40 , 41] . However , the three loci on large chromosome are likely to be considered the best for estimating across large distances [35] . Thus , when we considered the MLVA profiles in terms of the three loci ( VC0147 , VC0437 , and VC1650 ) on large chromosome , the isolates with profiles 8 , 4 , 6 , ( 50% , 39/78 ) and 9 , 4 , 6 , ( 10% , 8/78 ) were related to the Haiti and Bangladesh strains , respectively [11 , 35] . Our WGS analysis demonstrates that these Mozambican isolates formed a distinct lineage within a clade that includes El Tor variants from Bangladesh , China , Haiti , Nepal , India , and Kenya that belong to the current global radiation of the Seventh Pandemic . The Manhiça , Mozambique isolates differ from the wave 3 backbone by 39 nucleotides . Furthermore , previous reports demonstrated Wave 2 strains in Mozambique during the same time period [21 , 42] . Taken together with our analysis , it is clear that there were multiple , independent V . cholerae lineages from Wave 2 and Wave 3 which were circulating within Mozambique during this period of time . In Manhiça , although the relationships between isolates differed in detail between MLVA and WGS , both analyses demonstrated the isolates were very closely genetically related , even if they are collected several years apart .
Our findings further define the molecular epidemiology of V . cholerae in Mozambique . Our study demonstrates that Wave 3 isolates of the Seventh Pandemic have become established in Manhiça , Mozambique and have persisted in this region over the time of this study alongside V . cholerae Wave 2 strains in other regions of the country . The subsequent radiation of genotypes has been enriched by the process of recombination as detected in the WGS data . Our data raises several important questions that relate to where these V . cholerae isolates persist and how seemingly identical isolates can be collected years apart despite our understanding of the high rate of change of the MLVA loci and the V . cholerae molecular clock . Although the environmental triggers for the emergence of cholera are unknown in Manhiça , it is important to be vigilant to prevent an emergence from becoming an outbreak .
|
Cholera is a deadly disease caused by the bacterium Vibrio cholerae . The ancestral home of cholera is around the Bay of Bengal , but recently cholera has moved to Africa . In Africa , cholera occurs in sporadic outbreaks . In order prevent cases of cholera , we want to understand the transmission of cholera in Africa , does it stay in one place or does it move around . To gain insight into these questions , we have examined the DNA of the bacteria . The DNA provides a identity for each isolate and we can infer how the isolates are related to each other based on the number and type of DNA changes . In our study , we examined the DNA of cholera isolates from southern Mozambique . We were surprised how similar all the Mozambique isolates were even though that were collected up to eight years apart . Based on previous work , we would have expected much more change in the DNA . Our data raises several important questions that relate to where these cholera isolates persist , possibly in local refuges , and how seemingly identical isolates can be collected years apart despite our understanding of the high rate of change of the molecular clock .
|
[
"Abstract",
"Introduction",
"Methodology",
"Results",
"Discussion",
"Conclusion"
] |
[
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"rivers",
"pathogens",
"vibrio",
"variant",
"genotypes",
"tropical",
"diseases",
"microbiology",
"geographical",
"locations",
"genetic",
"mapping",
"bacterial",
"diseases",
"phylogenetics",
"vibrio",
"cholerae",
"data",
"management",
"aquatic",
"environments",
"bodies",
"of",
"water",
"neglected",
"tropical",
"diseases",
"bacteria",
"bacterial",
"pathogens",
"africa",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"cholera",
"chromosome",
"biology",
"medical",
"microbiology",
"marine",
"and",
"aquatic",
"sciences",
"microbial",
"pathogens",
"evolutionary",
"systematics",
"genetic",
"loci",
"people",
"and",
"places",
"mozambique",
"freshwater",
"environments",
"cell",
"biology",
"heredity",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"organisms",
"chromosomes"
] |
2017
|
Minimal genetic change in Vibrio cholerae in Mozambique over time: Multilocus variable number tandem repeat analysis and whole genome sequencing
|
Genetic transformation is a potential tool for analyzing gene function and thereby identifying new drug and vaccine targets in parasitic nematodes , which adversely affect more than one billion people . We have previously developed a robust system for transgenesis in Strongyloides spp . using gonadal microinjection for gene transfer . In this system , transgenes are expressed in promoter-regulated fashion in the F1 but are silenced in subsequent generations , presumably because of their location in repetitive episomal arrays . To counteract this silencing , we explored transposon-mediated chromosomal integration of transgenes in S . ratti . To this end , we constructed a donor vector encoding green fluorescent protein ( GFP ) under the control of the Ss-act-2 promoter with flanking inverted tandem repeats specific for the piggyBac transposon . In three experiments , free-living Strongyloides ratti females were transformed with this donor vector and a helper plasmid encoding the piggyBac transposase . A mean of 7 . 9% of F1 larvae were GFP-positive . We inoculated rats with GFP-positive F1 infective larvae , and 0 . 5% of 6014 F2 individuals resulting from this host passage were GFP-positive . We cultured GFP-positive F2 individuals to produce GFP-positive F3 L3i for additional rounds of host and culture passage . Mean GFP expression frequencies in subsequent generations were 15 . 6% in the F3 , 99 . 0% in the F4 , 82 . 4% in the F5 and 98 . 7% in the F6 . The resulting transgenic lines now have virtually uniform GFP expression among all progeny after at least 10 generations of passage . Chromosomal integration of the reporter transgenes was confirmed by Southern blotting and splinkerette PCR , which revealed the transgene flanked by S . ratti genomic sequences corresponding to five discrete integration sites . BLAST searches of flanking sequences against the S . ratti genome revealed integrations in five contigs . This result provides the basis for two powerful functional genomic tools in S . ratti: heritable transgenesis and insertional mutagenesis .
Parasitic nematodes have an enormous impact on human welfare , infecting over a billion people and causing debilitating , disfiguring or blinding disease in hundreds of millions [1] , [2] , [3] . More insidious effects of these parasites include complications in pregnancy and physical and cognitive deficits in children [4] , [5] . Parasitic nematodes severely degrade the health of domestic animals , bringing about significant economic losses in developed agricultural production systems [6] , [7] and heighten food insecurity in marginal economies [8] . The need for ongoing research into new agents to prevent or treat parasitic nematode infection is acute . There are currently no effective vaccines available for these parasitisms , and chemotherapy is based on a relatively small arsenal of anthelmintic drugs [9] , [10] . Resistance to each of these is widespread among parasites of livestock [11] , [12] , and suboptimal anthelmintic treatment responses in human clinical settings may signal genetically based resistance arising in populations of nematode parasites of humans [13] , [14] , [15] , [16] . Screening has and will likely continue to identify new candidate anthelmintics against nematodes . Increasingly , however , alternative approaches involving rational design of agents directed at defined molecular targets are important in developing new drugs and identifying potential vaccine candidates in other infectious disease systems and have recently been brought to bear on parasitic nematodes [17] . Burgeoning descriptive genomic and transcriptomic resources for parasitic nematodes notwithstanding [18] , [19] , [20] , [21] , [22] , [23] , functional characterization and validation of such molecular targets in these worms has been hampered by the lack of robust functional genomic tools such as transgenesis and targeted gene silencing or disruption . For this reason our laboratory has worked towards a system for transgenesis in parasitic nematodes of the genus Strongyloides [24] , [25] , [26] , [27] . We focus on Strongyloides spp . because unlike most parasitic nematodes , which display an invariant pattern of development to infective larvae outside the host , these worms execute one or more generations of free-living development between parasitic generations [28] . The free-living females of Strongyloides constitute an advantageous point of attack for gene transfer because they may be cultured in vitro and because morphological similarities between them and hermaphrodites of the free-living nematode Caenorhabditis elegans have made it relatively straightforward to adapt the technique of gonadal microinjection , originally devised for transgenesis in C . elegans [29] , [30] , [31] , to these parasites [27] . This technique , along with the design of vector constructs containing both 5′ and 3′ regulatory sequences from Strongyloides sp . has enabled a robust system for generating transgene expressing S . stercoralis larvae of the F1 generation following gene transfer [24] , [25] . In an effort to establish stable transgenic lines , F1 transgenic larvae of S . stercoralis have been reared to infective third-stages and used to establish patent infections in gerbils [24] , [25] . Transgene expression has been observed in parasitic females recovered from the intestines of these animals . Recently we demonstrated that transgene constructs containing regulatory sequences from S . stercoralis are expressed at roughly equal frequencies and in virtually identical anatomical patterns in Strongyloides ratti [26] . When F1 transgenics of S . stercoralis [24] , [25] and S . ratti ( unpublished ) transformed with conventional plasmid vectors are subjected to host passage , a substantial proportion of their F2 progeny harbor transgene sequences , but these are not expressed in this or subsequent generations of passage . We assume that like C . elegans , Strongyloides spp . assemble the majority of microinjected transgenes into highly repetitive episomal arrays [30] , [31] , and we hypothesize that Strongyloides spp . actively silence these because of their episomal location , their highly repetitive character [32] or both . We have addressed both of these scenarios by attempting low-copy integration of transgenes into the chromosomes of S . ratti using the piggyBac transposon system [33] . As an additional precaution against epigenetic silencing , we assessed the ability of the gypsy retroviral insulator sequences from Drosophila [34] to sustain transgene expression during host passage . Preliminary experiments with this system yielded the first transgene expressing individuals of the F2 generation in S . stercoralis and S . ratti observed to date [33] . In the present study we confirmed this finding in S . ratti , demonstrated that the piggyBac system results in chromosomal integration of transgenes , and we established lines of this parasite that stably transmit and express integrated transgenes in virtually all progeny .
To achieve low-copy integration of transgenes in Strongyloides we designed constructs that incorporate regulatory elements of the piggyBac transposon system ( Fig . 1 ) . The transgene encoded in both the donor constructs tested ( Fig . 1A , B ) included a previously reported GFP expression cassette in which expression is directed to the body wall by the promoter for the cellular actin gene Ss-act-2 [24] . In both donor constructs , this reporter transgene was flanked by the inverted terminal repeats specific for the piggyBac transposon [35] as well as internal transposon sequences shown to be necessary for efficient transposition in Drosophila [36] . In construct pPV356 ( Fig . 1B ) , the reporter transgene is also flanked by the gypsy retroviral insulator sequences as a further precaution against epigenetic silencing . It is worth noting that in an effort to define a minimal functional transposon for studies of gene function in insects [36] , the internal transposon sequences in the piggyBac encoding vector pXLBacII , from which we derived the piggyBac elements for our vectors , have been significantly reduced in length compared to the naturally occurring transposon from Trichoplusia ni [37] . Although this minimal transposon worked under the conditions obtaining in this initial study with S . ratti and mobilizes efficiently in both Trichoplusia [36] and Drosophila [38] , it is possible that its efficiency could be increased in future studies of Strongyloides and other parasites by optimizing the lengths of internal sequences in the donor vector . As discussed below , this issue may be particularly important in attempts to establish stable lines of dioecious parasitic nematodes with integrated transgenes . Function of the piggyBac transposase was encoded either in a second plasmid vector , the helper ( pPV402 , Fig . 1C ) , in which the enzyme was expressed ubiquitously [24] under the promoter for the ribosomal small subunit gene Ss-rps-21 , or in capped and tailed synthetic mRNA transcribed in vitro under the T7 promoter from linearized plasmid pPV257 ( Fig . 1D ) . In preliminary experiments [33] , transforming S . stercoralis and S . ratti with pPV356 along with mRNA encoding the piggyBac transposase resulted in efficient production of F1 transgenics in both parasites , and , following passage of these through gerbils or rats , respectively , the first transgene-expressing F2 parasites observed to date . Given the very low numbers of F2 transgenic S . stercoralis produced relative to the approximately 50 third-stage larvae required to establish a patent infection in the gerbil or dog , we focused the present study on S . ratti . Transgenes with regulatory elements from S . stercoralis are expressed in virtually identical patterns and at equal frequencies in S . ratti [26] . Most importantly , the availability of a well-adapted laboratory host , the rat , makes it possible to establish patent infections with as few as one or two infective larvae [26] , [33] . Transformation of S . ratti with pPV356 in the absence of transposase-encoding helper constructs gave efficient transmission and expression of transgenes in F1 progeny , but passage of these through rats failed to yield any transgene-expressing larvae in the F2 generation ( Table 1 , Experiment 1 ) . Transformation of S . ratti with donor plasmid pPV356 combined with capped RNA encoding the piggyBac transposase gave efficient transgene transmission and expression in F1 larvae , and it resulted in a small proportion of F2 individuals expressing the transgene ( Table 1 , Experiment 2 ) . This finding is consistent with our hypothesis that silencing observed in Strongyloides spp . transformed with plasmid vectors stems from the assembly of the encoded transgenes into episomal arrays and that this silencing can be circumvented by integration of the transgenes into the chromosomes of the parasites . Replicate experiments ( Experiments 3 and 4 , Table 1 ) , in which worms were transformed with pPV356 and the helper plasmid pPV402 , yielded markedly higher proportions of F2 parasites expressing the reporter transgene following host passage than seen when the transposase activity was encoded in capped RNA . This suggests that in S . ratti , co-transfected helper plasmids designed for in situ expression of the piggyBac transposase constitute a more efficient means of providing this activity than in vitro translated capped RNA . By contrast , co-transfection with capped RNA encoding the transposase gives efficient piggyBac-mediated integration of transgenes in Schistosoma mansoni , whereas expression of the transposase from co-transfected helper plasmids does not [39] . Co-transformation of S . ratti with donor plasmid pPV254 , which lacks the gypsy insulator sequences , and helper plasmid pPV402 ( Table 1 , Experiment 5 ) yielded transgene-expressing individuals in the F2 generation . This indicates that although the gypsy insulator sequences effectively resist positional silencing of transgenes in Drosophila [34] , they are not required for sustained transgene expression in S . ratti . This result further supports the hypothesis that chromosomal integration of transgenes is the essential factor provided by this system . Having observed the first transgene expression in the F2 generation of S . ratti , we asked whether these F2 individuals could found stable transgenic lines . We did so using the protocol illustrated in Figure 2A . Briefly , parental free-living females of S . ratti were transformed by gonadal microinjection with Ss-act-2::gfp in piggyBac vectors as described previously for plasmid vectors [24] , [25] . F1 progeny of microinjected females were reared in culture and screened for reporter expression by fluorescence stereomicroscopy . GFP+ progeny were cultured to infective L3 ( L3i ) and inoculated into rats . F2 progeny arising in feces of these rats were screened for GFP expression and hand selected . These F2 individuals were cultured to mating pairs of free-living adults , and their progeny ( F3 ) were reared to L3i , selected for expression of the transgenes , and inoculated into rats . S . ratti eggs and first-stage larvae in the feces of these rats constituted the F4 generation . This pattern of alternating culture , selection for expression and host passage was then repeated for subsequent generations of selection . The results ( Fig . 2B ) , expressed as the percentage of GFP+ individuals in the population screened within each generation , demonstrate derivation of stable lines with virtually 100% inheritance and expression of transgenes within 5 generations of selection . Numbers of individuals available for passage in the F1 and F2 generations were low for all three lines , but could be amplified owing to the high susceptibility of laboratory rats to S . ratti infection . The decline in the proportion of GFP-positive larvae in the F5 generation is noteworthy . In our selection scheme ( Fig . 2A ) , the F5 generation arises from crossing of GFP-positive free-living males and females in culture . It is possible that some heterozygous individuals remain in the F5 generation and that the observed decline in the frequency of GFP-expressing individuals in the F5 resulted from segregation of homozygous wild-type individuals among the progeny of crosses between heterozygotes or between heterozygotes and homozygous transgenic worms . We are continuing to maintain these lines by serial passage . Line PV2 is currently in the F10 and exhibits GFP fluorescence in 100% of individuals; PV3 is in the F10 with expression in 98 . 6% of individuals , and PV4 is in the F8 with expression in 99 . 4% of individuals . While expression rates in PV3 and PV4 may reflect remaining heterozygosity in these lines , they could also be due to diminished levels of expression or viability in small proportions of the worms examined by fluorescence stereomicroscopy . We have cryopreserved transgenic L3i from each line as described [40] ( at −80°C instead of in liquid nitrogen ) and have thawed viable late-passage worms with a recovery rate of approximately 10% ( data not shown ) . All thawed L3i were transgenic , as evidenced by expression of GFP , indicating that in the lines observed , integration of the transgenes did not affect parasite fitness in a way that would render them more susceptible to damage during cryopreservation . The expression pattern of the transcriptional reporter Ss-act-2prom::gfp::Ss-era-1 3′ in parasitic females from a stable integrated line ( PV2 ) of S . ratti were similar to that reported previously [24] for F1 larvae of S . stercoralis transformed with a conventional plasmid vector ( pAJ08 ) and presumably expressing this transcriptional reporter from a multi-copy episomal array . In general , parasitic females from line PV2 exhibited a zone of fluorescence in the body wall extending posteriorly from the junction between pharynx ( esophagus ) and intestine to a level approximately even with the posterior margin of the uterus ( Fig . 3A , B ) . Retention of the body-wall specific pattern of Ss-act-2-regulated GFP expression in stable S . ratti transformants is consistent with previous findings that promoters derived from S . stercoralis drive reporter transgene expression in identical or highly similar anatomical patterns in S . ratti [26] . Moreover , this observation suggests that expression patterns of transgenes integrated into limited numbers of chromosomal sites in Strongyloides spp . will be consistent with patterns resulting from over expression of the same transgenes from high copy number episomal arrays . The expected body wall specific pattern of the Ss-act-2prom::gfp reporter was also seen in free-living females ( Fig . 3C , D ) and free-living males ( Fig . 3E , F ) . In addition to uniform expression throughout the body wall , additional loci of GFP expression were seen in the vulva and rectum of free-living females ( Fig . 3D ) and in the cloaca of free-living males ( Fig . 3F ) . No fluorescent signal could be detected from non-transformed free-living male S . ratti ( Fig . 3G , H ) . Owing to the failure of non-integrated transgenes to express beyond the F1 generation , free-living adult Strongyloides expressing transgenes had not been observed prior to the establishment of stable integrated lines in S . ratti . We inferred chromosomal integration of the reporter transgene in S . ratti from a Southern blot of a BsrGI restriction digest of genomic DNA from each transgenic line which was hybridized with a gfp-specific probe ( Fig . 4A ) . We detected gfp-specific sequences in multiple fragments with various molecular weights in the genomic DNA of each line ( Fig . 4B ) . Based on the presence of only one BsrGI site in the vector , occurring within the gfp coding sequence , the smallest predicted transgene-containing restriction fragments of the genome would be 1898 bp for worms transformed with the donor vector pPV356 , which contains the 430 bp gypsy insulator sequences or 1468 bp for worms transformed with donor vector pPV254 , which does not contain these sequences ( Fig . 4A ) . The Southern hybridization patterns seen for PV2 , PV3 and PV4 ( Fig . 4B ) are generally consistent with this prediction , showing a multiplicity of bands hybridizing with the gfp-specific probe in the range of approximately 1 . 8 kb to 7 . 5 kb . By contrast , this diverse banding pattern is not consistent with the uniform 8084 and 6758 bp fragments that would be expected to result from BsrGI digestion of episomal arrays made up of the donor vectors pPV356 and pPV254 , respectively . Chromosomal integration was confirmed by splinkerette-PCR , which allowed sequencing of the genomic regions bounding unique transgene insertions in five contigs of the S . ratti genome ( Table 2 ) . The current draft genome assembly for S . ratti ( http://www . sanger . ac . uk/resources/downloads/helminths/strongyloides-ratti . html ) allows assignment of three of these contigs to Chromosome I . Further analysis , in which we subjected 2 kb sequences flanking each of the five identified transposon insertion sites ( Table 2 ) to BlastN/BlastX searching against the non-redundant nucleotide/protein sequences , revealed three insertions in coding regions . All insertions were at TTAA sites ( Table 2 ) as has been found in other studies of the piggyBac transposon [21] , [39] , [40] . Insertion PV2-2 ( Table 2 ) is near the 5′ end of a sequence predicted to encode a peptide that is 51% identical to the hydroxyacylglutathione hydrolase in Loa loa ( Accession XP_003137817 ) . Insertion PV3-1 is within the coding sequence of a 28S ribosomal RNA gene that has 98% nucleic acid identity with AB205054 in S . stercoralis , and insertion PV4-1 is within a gene sequence predicted to encode a peptide with 68% identity to hypothetical protein T13G4 . 3 in Caenorhabditis elegans ( ADD13552 . 1 ) ( XP_001894192 . 1 ) . Insertions PV2-1 and PV2-3 were in intergenic regions . Although the wide molecular weight ranges of restriction fragments of gDNA with gfp-specific sequences in all three lines ( Fig . 4B ) are suggestive , the breadth of transgene integrations throughout the genome of S . ratti cannot be assessed accurately from the few sequenced integration junctions reported here . Additionally , splinkerette-PCR , the technique used to sequence and map integration boundaries , is biased to the restriction sites of the enzymes used to digest the genomic DNA . In future studies , high-capacity genome sequencing could be brought to bear , as done previously with Salmonella typhi [41] , to determine whether the pattern of piggyBac insertions in S . ratti is consistent with the observed broad patterns of piggyBac insertions in the genomes of Drosophila melanogaster [42] and the parasitic trematode S . mansoni [39] . The fact that three of the five insertions we sequenced were in coding sequences in the in S . ratti genome is consistent with the piggyBac transposon's documented propensity to integrate into the coding regions of genes in other organisms [42] , [43] . This observation underscores the potential of the piggyBac transposon system as a tool for insertional mutagenesis in unbiased forward genetic investigations with S . ratti . Estimates of relative copy number of integrated transgenes derived by qRT-PCR were similar in the three independent lines with mean relative copy number ranging from 37 to 51 per genome ( Table 3 ) . These means were not significantly different ( P>0 . 05 ) , indicating that the transformation protocol results in a consistent number of transgene integrations . To ensure that the estimates of relative copy number of integrated transgenes did not reflect sequences in episomal arrays , we conducted PCR on gDNA from each stable line using forward and reverse primers hybridizing within the M13 reverse priming site of the vector and within the transposon sequence respectively . These primers did not yield a product from gDNA of the stable integrated lines ( F8 generation ) , but did from the F1 progeny of free-living female worms microinjected with donor ( pPV356 ) and helper plasmids ( Fig . 4C ) . This suggests that donor plasmids are transmitted to the F1 generation but are lost during passage . We conclude , therefore , that our estimates of relative transgene copy number reflect numbers of chromosomal integrations . While random integration of large numbers of transposons would be desirable in unbiased studies aimed at gene tagging or disruption , integration of large numbers of transgene copies in focused studies of gene function is less than desirable given the risk of non-physiologic effects of over expression and non-specific effects resulting from insertional mutagenesis . It is noteworthy in this regard that no unusual or unexpected phenotypes have been evident among worms from the three stable lines in the present study . A technique for single-copy integration of transgenes in C . elegans based on the Mos-1 transposon system has been developed recently [44] and could provide a strategy for achieving similar results in Strongyloides spp . in the future . Note that a stable , GFP-expressing line was established using the donor plasmid pPV254 , which lacks the gypsy insulator sequences , ( Fig . 1; Tables 1 and 2 ) . Furthermore , the proportion of GFP expressing individuals among all larvae screened did not vary significantly as a function of donor construct ( Fig . 2B ) . Together , these observations indicate that under conditions pertaining in the present study , where integrations appeared to occur widely throughout the genome , with mean relative copy numbers ranging from 37–51 per genome , the gypsy insulator sequences are not required for stable expression of transgenes in integrated lines of S . ratti . However , the gypsy sequences were originally characterized as resisting positional silencing effects in Drosophila , which vary significantly based on the site of transgene integration [34] . Although positional silencing of transgenes has not been documented in S . ratti , it is possible that , as we attempt to achieve integrations at lower copy number or to adapt methods for single copy transposon-mediated transgenesis [44] to this parasite , such positional effects will manifest themselves and be a significant factor in stable transgene expression . In such cases it may be prudent to include the gypsy sequences as data from the present study also indicate that they do not adversely affect the efficiency of integration . In summary , we have developed a transposon-based system for integration of transgenes into the chromosomes of Strongyloides spp . and for the establishment of stable transgenic lines . Strongyloides ratti lends itself particularly well to establishment of stable integrated lines owing to its efficiency of infection in laboratory rats . Three such integrated lines have been established in S . ratti , and boundary sequences of multiple unique transgene integration sites have been mapped . The possibility of introducing heritable , stably expressed transgenes into a parasitic nematode is an advance that offers important new opportunities to study gene function in these organisms . Perpetuation of stable transgenic lines will greatly facilitate functional approaches involving expression transgenes with dominant interfering or activating mutations as was recently done with very limited samples of transiently transformed S . stercoralis [45] . Transposon mediated transgenesis will also be a useful adjunct to other methods of genetic analysis such as RNAi . The majority of animal parasitic nematodes exhibit very limited sensitivity to exogenously administered RNAi [46] . RNAi targets in Haemonchus contortus that are located on cell surfaces in contact with the worm's external environment show greater RNAi sensitivity than internal targets [47] , suggesting that this , and perhaps other parasitic species , are deficient in mechanisms that spread administered dsRNA from cell to cell . Endogenously expressing interfering RNAs from transgenes in stably transformed worms could be a means of surmounting barriers to uptake and transport of administered dsRNA . This approach has shown promise in transiently transformed schistosomes [48] , [49] , [50] . Furthermore , by stable heterologous expression of dsRNA transporters such as C . elegans SID-2 [51] or of Argonaut proteins such as RDE-4 , it may be possible to complement specific deficiencies in RNAi processing suggested by recent transcriptomic studies in parasitic nematodes [33] , [52] . With regard to the use of a transposon-based approach to transgene integration per se , improvements in the efficiency of transposition will enable such techniques as transposition induced gene knockouts [42] , exon and enhancer traps [53] , and excision-based site directed mutagenesis [54] to further the genetic analysis of these species . The use of transposon integration sites as genetic markers [55] could aid in the assembly of chromosome scaffolds for the ongoing genome sequencing projects for these worms . While the present finding constitutes a substantial advance in transgenesis for Strongyloides and related genera , its impact would be greater if the technology could be adapted to other parasitic nematodes of medical or agricultural importance . We envision two potential hurdles , the lack of parthenogenetic parasitic females and challenges in gene transfer , that would need to be surmounted in order to accomplish this . The property of parthenogenesis in parasitic females of Strongyloides spp . [28] represents an advantage in deriving transgenic lines by host passage in that it makes it possible to obtain large numbers of progeny from parasitic females without the presence of males . Moreover , progeny from individual parthenogenetic females are clonal [56] and so matings in subsequent free-living generations are effective selfings of the parasitic female parent , further hastening the selection of homozytotic transgenic parasites . These advantages would not be available in other parasitic nematode taxa of interest , which are dioecious . Nevertheless , if sufficient numbers of transgenic progeny from the F1 and subsequent generations , comprising both male and female larvae , and a sufficiently well adapted host-parasite system in which to carry out line selection were available , these conditions could combine to enable the establishment of stable transgenic lines of dioecious parasitic nematodes , albeit with a more lengthy and labor intensive process than in Strongyloides spp . The ability to transfer genes into Strongyloides spp . and related genera by gonadal microinjection of free-living females [27] , [57] also constitutes a major advantage in undertaking transgenesis that will be unavailable in the majority of parasitic nematode species that cannot undertake full generations of free-living development . However , larval stages of animal parasitic nematodes that arise in the extrinsic environment or in arthropod vectors can be cultured transiently and may be susceptible to gene transfer by other routes . This potential was recently demonstrated emphatically in the heritable transformation of Brugia malayi by chemically mediated gene transfer [58] . Gene transfer in this study was achieved by incubating developing third-stage larvae with co-precipitates of the DNA vector with calcium phosphate . This method succeeded with larvae that were inoculated along with the co-precipitate into the peritoneal cavities of susceptible gerbils but not with larvae cultured in vitro with co-precipitate , even in a culture system that promoted L3–L4 molting . This finding may indicate that completion of a molt cycle with subsequent remodeling of the nascent cuticle , which occurs in the host , but may not occur normally in cultured B . malayi L3 , is necessary for sufficient uptake of the DNA-CaP04 co-precipitate . In any case , this method of chemically mediated gene transfer into an accessible larval stage represents one that , in theory , could be adaptable to transposon mediated integration of transgenes into a broad range of parasitic nematodes of medical and veterinary importance .
Rats and gerbils were used for parasite strain maintenance and host passage of transgenic S . ratti in this study . These procedures were approved by the Institutional Animal Care and Use Committee ( IACUC ) of the University of Pennsylvania ( Protocol No . 803511 ) . This protocol , as well as routine husbandry care of the animals , was in strict accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The ED321 strain of S . ratti was maintained in rats and gerbils and cultured as described [59] . Free-living adult S . ratti were isolated via the Baermann funnel technique from charcoal coprocultures maintained at 22°C for 48 hours . The worms were washed twice with sterile deionized water to reduce carryover of fecal bacteria and plated on Nematode Growth Medium ( NGM agar ) plates with lawns of Escherichia coli HB101 . Agar plate cultures of S . ratti were incubated at 22°C unless otherwise noted . Rats were infected with S . ratti by subcutaneous injection of infective larvae ( L3i ) . Naïve rats are highly susceptible with S . ratti , frequently developing patent infection after inoculation of only one or two infective larvae [60] , [61] . This makes them the host of choice for early generations of passage during isolation of stable lines , when very few transgenic L3i are available . Rats , however , mount a vigorous protective immune response to S . ratti infection , and eliminate their infections about one month following inoculation [62] . Gerbils are also susceptible to infection with S . ratti but require 50–100 L3i to establish a patent infection . However , gerbils do not expel their infections as rapidly as rats and so allow maintenance for 6 months or more in individual animals [63] . This makes them the hosts of choice for maintaining established stable transgenic lines , where numbers of L3i are not limiting . Therefore , once established by 3 or 4 generations of alternating passage through free-living culture , selection and rat infection , stable transgenic lines of S . ratti were maintained by serial passage in gerbils thereafter . Cohorts of L3i from each stable line were also cryopreserved as described [40] and the viability of thawed parasites verified by observation of patency following gerbil inoculation . We created donor vector pPV254 ( Fig . 1A; GenBank accession number , JX013636 ) by excising the previously described reporter cassette , Ss-act-2p::gfp::Ss-era-1 3′ UTR , from pAJ08 [24] ( Addgene Plasmid #14912 ) with HindIII and EagI ( Unless otherwise noted , all restriction and other DNA processing enzymes used in this study were obtained from New England Biolabs ) and ligating with T4 DNA Ligase into the piggyBac donor vector pXL-BacII [38] containing inverted terminal repeats ( ITRs ) of the piggyBac transposon and essential internal sequences surrounding a multi-enzyme cloning site ( MCS ) , which was cut with the same restriction enzymes . As a precaution against epigenetic silencing , the gypsy retroviral insulator sequences , which retard positional silencing effects in Drosophila [34] , were introduced at the 5′ and 3′ ends of the expression cassette in donor vector pPV254 to create the new donor vector pPV356 ( Fig . 1B; GenBank accession number , JX013635 ) . Tandem 430 bp gypsy insulators were excised from plasmid pCa4B2G , a kind gift from Norbert Perrimon , Harvard University School of Medicine , by digestion with Sac I and Not I . The fragment containing the gypsy insulators was isolated by gel purification and blunt ended by Pfu turbo DNA Polymerase ( Stratagene ) . Plasmid pPV254 was then digested with Hind III and Xba I , and the larger ( 3970 bp ) fragment , which contained the piggyBac ITRs but lacked the expression cassette , was isolated and blunt ended as described above . These two fragments were then ligated to yield a piggyBac plasmid with two gypsy insulators . Subsequently , a multi-cloning site ( MCS ) ( GGGCCCACTAGTCGGCCGGAATTCCTAGGACCGGTACCCTGCAGAAGCT ) was cloned into the plasmid between the gypsy insulators to create a multi-purpose insulated piggyBac vector . Finally , the expression cassette , Ss-act-2p::gfp::Ss-era-1 3′UTR from pAJ08 was cloned into the MCS by digestion with Xba I and Hind III and ligation with T4 DNA Ligase to yield pPV356 . pPV402 ( Fig . 1C; GenBank accession number , JX013634 ) , the piggyBac transposase helper vector , was made by excising the piggyBac transposase coding sequence from pPV257 ( see below ) with the restriction enzymes AgeI and AvrII and cloning them into the pAJ50 vector [24] ( Addgene Plasmid #14918 ) from which the gfp coding region had been removed with the same restriction enzymes . This yielded pPV402 , in which the piggyBac transposase is expressed under the control of the Ss-rps-21 promoter , which drives expression of transgenes in all tissues including the germline of microinjected P0 female worms and in F1 transformants [24] . To prepare in vitro transcribed synthetic mRNA for microinjection in lieu of a helper plasmid , pPV257 ( Fig . 1D; GenBank accession number JX017375 ) was made by first amplifying the piggyBac transposase coding sequence from pBSII-IE1-orf ( http://piggybac . bio . nd . edu/ ) with the PCR primers HelpKpnIAgeIF ( CCTGCAGCCCGGGTACCGGTATATAATAAAATGGG ) and HelpAvrIIEcoRIR ( GTGGCGGCCGCTGAATTCCTAGGGGATCCAAATTC ) using the Pfu Turbo DNA polymerase ( Stratagene ) . The PCR amplicon was cloned into the pCR-Blunt II-TOPO vector ( Invitrogen ) to yield pPV257 . This construct was linearized with AvrII and used for in vitro transcription from the T7 promoter of the parental pCR-Blunt II-TOPO plasmid using the mMessage mMachine T7 Ultra kit ( Ambion , Austin , TX , USA ) , which includes capping and poly-A tailing reactions , according to the manufacture's instructions . Constructs were microinjected into one arm of the syncytial gonad of free-living S . ratti females as described for C . elegans [31] . Microinjected worms were recovered on NGM ager plates with lawns of Escherichia coli HB101 and cultured at 22°C with free-living males . At 48 h following injection , parental ( P0 ) females and their broods were observed , and both total and GFP-positive F1 eggs and larvae were counted . Over the ensuing week , GFP positive third stage larvae ( L3i ) of the F1 generation were reared , manually selected and stored in BU buffer [64] for animal inoculation . Basic steps of this procedure and their timing are illustrated in Fig . 2A . Frequencies of GFP expression in worms screened from each of three transgenic lines , including one transformed with donor vector pPV254 and two transformed with pPV356 , were compared statistically as detailed in the caption to Fig . 2 . The criterion for significance was P≤0 . 05 . Genomic DNA was extracted from control and transformed S . ratti from the F8 generation of each stable line using the Gentra Puregene Tissue Kit ( QIAGEN ) . Genomic DNA and donor plasmid pPV356 were digested with BsrGI . Each donor vector , pPV254 and pPV356 , contains only one BsrGI site . BsrGI was selected because the location of its single cut site within the gfp coding sequence would result in two fragment ends that hybridize with the probe for each integration event , thus increasing the intensity of the signal in our Southern blot . Restriction fragments were separated by electrophoresis through 1% agarose , transferred to nylon membranes and fixed there by cross linking with UV light using a UV Stratalinker 1800 ( STRATAGENE ) . A probe homologous to a 723 bp segment of the gfp coding sequence was generated by PCR using primers: GFP-metF ( 5′ ATGAGTAAAGGAGAACTTTTC 3′ ) and GFP-702R ( 5′ ATCGCCAATTGGAGTATTTTGT 3′ ) . The PCR product was purified by agarose gel electrophoresis followed by gel extraction with the QIAquik kit ( Qiagen ) . The probe was labeled with digoxigenin–dUTP using DIG High Prime DNA labeling and detection starter kit II ( Roche ) . The DIG-labeled gfp gene probe was hybridized to Southern blots at 42°C overnight with gentle agitation . Membranes were washed at high stringency , the probe detected immunologically with CSPD and the resulting blot imaged on X-ray film ( Kodak ) . Chromosomal integrations of transgenes were further confirmed and mapped by splinkerette-PCR ( spPCR ) using a modification of Splinkerette Protocol S1 [65] . Genomic DNA was extracted from free-living adult worms of the transgenic lines using the Gentra Puregene Tissue Kit ( QIAGEN ) . Purified DNA ( ∼100 ng ) was digested with BstYI , BamHI or Bgl II for 2 h in a total volume of 15 µl for each reaction . Digested DNA was ligated with T4 Ligase to annealed splinkerette oligonucleotides ( SPLNK-GATC-TOP and SPLNK-BOT in protocol S1 ) with GATC sticky ends for 8 h at 16°C in a total volume of 30 µl . Following ligation of gDNA restriction fragments to splinkerette oligos , those containing transgene integrations were identified , and integration boundaries recovered by primary and nested spPCR using appropriate combinations of primers Splnk-1 and Splnk-2 , which target the ligated splinkerette oligos , and 3′PB1 and 5′PB1 , which target the 3′ and 5′ ends of the transgene coding sequence , respectively . For primary spPCR , Splnk-1 was paired with either 3′PB1 or 5′PB1 , and amplification was carried out with Phusion Polymerase and approximately 20 ng of splinkerette-ligated genomic DNA as template . Thermal cycling conditions were 98°C for 1 min followed by 30 cycles of 98°C for 20 sec , 55°C for 15 sec and 72°C for 2 min , with a final extension at 72°C for 10 min . Nested PCR was carried out with Splnk-2 paired with 3′PB2 or 5′PB2 and a 1∶20 dilution of spPCR products as template . Amplification with Phusion Polymerase was carried out using the same thermal cycling conditions detailed for primary spPCR . Primary and nested PCR products were analyzed by 1% agarose gel electrophoresis . After treatment using Shrimp Alkaline Phosphatase ( United States Biological ) and Exonuclease I , nested PCR products were sequenced using primers , 3′PB SEQ and 5′PB SEQ , targeting the 3′ and 5′ ends of the transgene coding sequence , respectively . Relative numbers of transgene copies integrated into each of the stable lines were estimated by real-time PCR using gDNA as template as described [66] . Relative copy number was expressed as the mean of Ct ( 2−ΔΔCt ) determined from three independent amplifications . Transgene DNA was quantified using primers specific for the coding sequence of gfp: gfp-F ( 5′ACCCTTGTTAATAGAATCGAG3′ ) and gfp-R ( 5′TCAATGTTGTGTCTAATTTTGAAG3′ ) . The gene aap-1 , which encodes the ortholog of the phosphoinositide 3-kinase ( PI3K ) p50/p55 adaptor/regulatory subunit , exists as a single copy in S . ratti genome and so was selected as a reference gene for relative copy number determination . Primers aap-1F ( 5′TACCAGAAGATGATGTAGATGC3′ ) and aap1R ( 5′AGTTTATTGACTTTAGTTGTCAATG3′ ) were designed based on the sequence of S ratti aap-1 . The size of amplicons from the aap-1- and gfp-specific primers were identical at 210 bp . Real-time PCR amplification was performed with a 7500 Fast Real-time PCR system using the SYBR Green Master Mix ( Applied Biosystems , Foster City , CA , USA ) . 10 ng of genomic DNA template were added in a total volume of 20 µl for each reaction . Each sample was set up in triplicate to give three technical replicates of the reaction . Reaction conditions were: start at 50°C for 2 min , initial denaturation at 95°C for 10 min , followed by 40 cycles of 95°C for 15 sec , 56°C for 1 min . The dissociation curve was run at 56°C for 60 min . Three independent amplifications ( denoted “measuments” in Table 3 ) were performed on gDNA from each line with three technical replicates per amplification . As stipulated previously [66] , quality control criteria for reliable results were a Ct<25 for an amplicon and a standard deviation <0 . 3 for technical replicates in an amplification . Relative copy number determinations for the integrated transgenes could be subject to inteference from transgene sequences in extrachromosomal arrays formed by the donor vector . To assess this eventuality , we conducted diagnostic PCR reactions with primers designed to detect vector sequences , presumed to be in extrachromosomal arrays , persisting in worms from the three stable transgenic lines . We preformed PCR using a pair of donor vector-specific primers for diagnostic PCR: M13 reverse , hybridizing with the M13 reverse priming site in the vector , and 5′PB2 ( see splikerette PCR protocol above ) hybridizing within the piggyBac transposon sequences of both vectors . These primers amplify an identical 624 bp product from both donor vectors used in the study . The primers SrAct-2F ( 5′-TGGAGATGAGGCCCAATCC-3′ ) and SrAct-2R ( 5′-GTGATGAAGATGAAGCAGCTGTG-3′ ) were designed to amplify a 554 bp fragment of endogenous gene Sr-act-2 , which was used as loading control . The DNA templates in the amount of 5 ng were used in the PCR reaction , which was carried out with Phusion Polymerase . Thermal cycling conditions were 98°C for 1 min followed by 25 cycles of 98°C for 20 sec , 57°C for 15 sec and 72°C for 40 sec , with a extension at 72°C for 10 min . Screening for transgenic larvae based on GFP fluorescence was carried out with a Olympus SZX12 stereomicroscope equipped with coaxial epifluorescence . Fine-scale examination of transgene expressing worms was carried out with an Olympus BX60 compound microscope with Nomarski Differential Interference Contrast ( DIC ) optics and epifluorescence ( Olympus America Inc . , Center Valley , Pennsylvania , USA ) . Images from the Olympus BX60 were captured with a Spot RT Color digital camera and processed using either the Spot Advanced image analysis software package ( Diagnostic Instruments , Inc . , Sterling Heights , Michigan , USA ) or Adobe Photoshop 7 . 0 . Image-processing algorithms such as contrast and brightness adjustments were always applied in linear fashion across the entire image .
|
Parasitic roundworms sicken and debilitate over one billion people , most of whom subsist on less than two US dollars per day . There are no vaccines and few drugs available to treat and prevent these infections . Basic research leading to new therapies has been hampered because we lack methods to study gene function in parasitic roundworms . One such method is transgenesis , a process by which gene function is inferred by studying the effects of transferring native or altered copies of genes into subject organisms . Our laboratory has developed a system for transferring synthetic genes into parasitic roundworms of the genus Strongyloides and for obtaining temporary expression of these “transgenes” . Until now , however , we have been unable to propagate these transgenic parasites in the laboratory . This paper describes a new technique that allows us to establish and maintain self-perpetuating lines of transgenic parasites for study . This represents a fundamental advance in the methodology for studying gene function in parasitic roundworms and should greatly facilitate the discovery of new therapies .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"infectious",
"diseases",
"molecular",
"cell",
"biology",
"global",
"health",
"genetics",
"biology",
"genomics",
"microbiology",
"genetics",
"and",
"genomics"
] |
2012
|
Transposon-mediated Chromosomal Integration of Transgenes in the Parasitic Nematode Strongyloides ratti and Establishment of Stable Transgenic Lines
|
The Skp1-Cul1-F box complex ( SCF ) associates with any one of a number of F box proteins , which serve as substrate binding adaptors . The human F box protein βTRCP directs the conjugation of ubiquitin to a variety of substrate proteins , leading to the destruction of the substrate by the proteasome . To identify βTRCP substrates , we employed a recently-developed technique , called Ligase Trapping , wherein a ubiquitin ligase is fused to a ubiquitin-binding domain to “trap” ubiquitinated substrates . 88% of the candidate substrates that we examined were bona fide substrates , comprising twelve previously validated substrates , eleven new substrates and three false positives . One βTRCP substrate , CReP , is a Protein Phosphatase 1 ( PP1 ) specificity subunit that targets the translation initiation factor eIF2α to promote the removal of a stress-induced inhibitory phosphorylation and increase cap-dependent translation . We found that CReP is targeted by βTRCP for degradation upon DNA damage . Using a stable CReP allele , we show that depletion of CReP is required for the full induction of eIF2α phosphorylation upon DNA damage , and contributes to keeping the levels of translation low as cells recover from DNA damage .
E3 ubiquitin ligases , which facilitate the attachment of anywhere from one to a long chain of the small protein ubiquitin to substrate proteins , are important regulators of the cell cycle and the response to stress . The best-studied outcome of ubiquitination is destruction of the substrate by the proteasome . There has been a great deal of interest in the discovery of ubiquitin ligase substrates , with the recent introduction of techniques that either look for proteins whose levels change when a particular ubiquitin ligase is inhibited [1–5] , or those that use mass spectrometry to look for proteins that interact physically with the ubiquitin ligase [6–11] . Unfortunately , some ligase-substrate interactions are likely too weak to purify by affinity . Moreover , once a list of associated proteins is identified , it is not always clear which are direct substrates . To address this , most studies have determined whether the half-life of the substrate is significantly altered upon inhibition of the ligase [11] . However , in many instances , only a select fraction of substrate is targeted . In addition , some substrates are targeted redundantly by multiple ligases [12] . These facts often make it impossible to verify candidates merely by examining their half-life . For ubiquitin ligases for which a consensus binding sequence is known , the presence of this sequence has been used frequently to separate true substrates from non-substrate or non-specific interactors . However , this method is not useful to discover substrates of the vast majority of ubiquitin ligases , for which no consensus sequence is known . To eliminate these problems , we developed a technique called Ligase Trapping [13] ( Fig 1A ) , in which an E3 ubiquitin ligase is fused to a ubiquitin-associated ( UBA ) domain . This mediates an extended interaction between the E3 ligase and its ubiquitinated substrates , allowing their co-immunoprecipitation . To distinguish between substrates and other associated proteins , this immunoprecipitate is subjected to a second purification for 6xHIS-ubiquitin under denaturing conditions . These purifications can be used both for substrate identification and as a diagnostic for candidate confirmation , in cases where the bulk level of a protein is stable . The SCF is a cullin-RING ligase ( CRL ) containing 3 core catalytic subunits: the RING finger protein RBX1 , the cullin CUL1 and the adaptor SKP1 [14–17] . This catalytic base associates with a substrate adaptor called an F box protein , of which humans encode at least 69 . F box proteins are thought to recognize their substrates only after substrate modification , typically by phosphorylation [14 , 17] . Several of these F box proteins have been characterized due to their well-established roles as tumor suppressors and oncogenes . βTRCP[18] is an F box protein that turns over substrates to control the G2/M transition ( e . g . WEE1 [19]/CDC25 [20 , 21] ) , as well as the response to DNA damage ( e . g . CDC25 [20 , 21] , claspin [7 , 22] ) . In this paper , we establish ubiquitin ligase trapping in mammalian cells . Of the 28 candidates identified using this technique , 12 were well-established substrates [6 , 20 , 21 , 23–33] . For the 16 remaining candidates , we examined 14 and found that 11 of these confirmed . Thus , 23 of the 26 known/tested candidates , ( 88% ) appear to be substrates , suggesting that Ligase Trapping is a robust discovery technique . Further characterization showed that turnover of one of the βTRCP substrates , CReP , is exacerbated by DNA damage . CReP is a protein phosphatase 1 ( PP1 ) specificity subunit that counteracts the phosphorylation of eukaryotic initiation factor 2 alpha ( eIF2α ) on serine-51 [34] , a stress-induced modification that inhibits translation initiation on most transcripts [35 , 36] . Inhibiting the turnover of CReP after DNA damage significantly reduced the accumulation of serine-51 phosphorylated eIF2α , and increased translation after DNA damage , suggesting that CReP turnover is an important mechanism by which DNA damage regulates translation .
To establish Ligase Trapping in human cells , we created a stable HEK293 line in which 6xHIS-ubiquitin is expressed upon treatment with doxycycline . In this cell line , tagged ubiquitin accounts for a significant portion of the total ubiquitin pool when cells are treated with doxycycline ( S1A Fig ) . In yeast , fusion of F box proteins , via a 3xFlag linker , to the UBA of Dsk2 or the two tandem UBAs of Rad23 , led to enhanced purification of nascent ubiquitinated F box protein substrates [13] . We fused the human F box protein βTRCP to the human homologs of these UBA-containing proteins , and found that the RAD23B fusion increased the poly-ubiquitinated species purified by the βTRCP fusion most strongly ( S1B Fig ) . Accordingly , we made a stable cell line that expressed both doxycycline-inducible 6xHIS-Ub and a Ligase Trap consisting of βTRCP fused on its C-terminus to 3xFlag and the C-terminal UBAs of RAD23B . To determine whether the βTRCP trap was functional , we expressed an epitope-tagged allele of the βTRCP substrate ATF4 in our stable cell line . We were able to immunoprecipitate poly-ubiquitinated ATF4 with the βTRCP trap , but not with the Ligase Traps of two unrelated F box proteins , FBXO24 and Fbw7 ( Fig 1B ) . We obtained a similar result with β-catenin ( S2 Fig ) . We also purified ubiquitinated forms of the Ligase Traps , which was unsurprising as many ubiquitin ligases are themselves ubiquitinated . We also purified substantial unmodified forms of the Ligase Traps . This is likely a result of the very large amount of IP loaded relative to input ( 5 , 000:1 for the 2nd step ) , which is necessary to see the very small percentage of substrate that is poly-ubiquitinated . Even in cases where the unmodified band is equal in intensity in the input and 2nd step IP , this represents only 0 . 02% IP background . This phenomenon also occurs frequently with unmodified substrates , while the relevant purification of poly-ubiquitinated substrates is highly specific to the relevant Ligase Trap . To examine further whether the purification of β-catenin was specific , we made a stable cell line identical to our βTRCP ligase trap line , but with a mutation in the WD40 domain of βTRCP predicted to prevent binding to β-catenin [37] . As expected , this mutant trap failed to purify polyubiquitinated β-catenin ( Fig 1C ) , showing that β-catenin purification by βTRCP represents a specific interaction . To make certain that the βTRCP Ligase Trap didn’t simply bind all ubiquitinated proteins more efficiently , we made a similar stable cell line expressing Fbw7-3xFlag-RAD23 . Poly-ubiquitinated forms of the known Fbw7 substrate MED13 [10] were preferentially precipitated with the Fbw7 Ligase Trap ( Fig 1D ) . Having established the functionality of the βTRCP ligase trap cell line , we performed a large-scale , two-step purification and identified ubiquitinated co-precipitating proteins by mass spectrometry . Before collection , we treated cells with the proteasome inhibitor MG132 for four hours , as we had shown that this treatment increases the amount of poly-ubiquitinated material purified by the βTRCP ligase trap ( S1C Fig ) . We defined candidate βTRCP substrates as those proteins identified in at least two of three purifications of the βTRCP ligase trap , but not in any of the negative control purifications . Twenty-eight proteins met these criteria ( Table 1 ) . Of these , twelve were previously-validated βTRCP substrates , and many others had been shown to interact with βTRCP in previously published large data sets , but had not been individually examined to determine if they were substrates [4 , 8 , 11 , 38–40] . SUN2 was purified in a large-scale screen for βTRCP substrates , and shown to be stabilized by the proteasome inhibitor MG132 [39] while this manuscript was under review . In addition , several other known βTRCP substrates , such as ß-catenin [41–45] , were selectively identified in the βTRCP purification , but as some peptides were also identified in control purifications , these did not meet the stringent criteria that we had chosen for this initial analysis ( bottom of Table 1 ) . The large fraction of previously-published substrates ( 43% ) that we purified confirms that Ligase Trapping accurately identified true substrates . We also purified substrates of Fbw7 using a Ligase Trap . The Fbw7 Ligase Trap was expressed at a low level , suggesting that this trap was less stable . However , the proteins pulled down most abundantly and specifically by the Fbw7 Ligase Trap were MED13 and MED13L , two members of the Mediator complex shown to be Fbw7 substrates in a recent screen[10] in which Fbw7 interactors were precipitated and identified by mass spectrometry . ( Our purification of MED13 is shown in Fig 1D ) In that screen , the entire 26-member Mediator complex was purified , and MED13 and MED13L had to be identified as the direct Fbw7 substrates by a combination of degron prediction and careful validation; we did not purify any other members of the Mediator complex . Ligase Trapping also provided a method to validate candidates beyond simply examining substrate turnover . Ligase Trapping is able to show that a ubiquitinated substrate specifically purifies with a particular ligase even if the substrate is redundantly targeted by multiple ligases , or if only a small fraction of the substrate ( such as that in a particular complex ) is ubiquitinated . To fully assay the accuracy of the Ligase Trap technique , we decided to validate candidate βTRCP substrates . Out of fourteen of the previously unknown/unvalidated candidates that we examined , eleven showed specific purification of polyubiquitinated material by the βTRCP ligase trap ( Table 1 and Figs 2 , S4 and S5 ) . This strongly suggested that these candidates are true substrates of βTRCP , and that this technique accurately identified substrates with low background and thus will be an efficient way of identifying and validating substrates of other ubiquitin ligases in the future . Two βTRCP candidate substrates were not examined due to technical difficulties . In order to determine whether βTRCP could bind its candidate substrates in the absence of the UBA domains present in the Ligase Traps , we co-expressed Flag-tagged versions of these F box proteins in HEK293 cells with HA-tagged versions of a subset of their candidate substrates . In all cases , the substrate was purified more efficiently by its cognate ligase than by the negative control ligase ( Fig 3A ) . Because a common outcome of ubiquitination by the SCF is proteasomal degradation of the ubiquitinated protein , we assayed whether a subset of the candidate substrates were degraded in a way that depended on the cognate ligase . For five of the βTRCP candidate substrates , we co-transfected cells with DNA encoding tagged substrate , as well as a negative control plasmid or a plasmid expressing an shRNA targeting both paralogs of βTRCP , then inhibiting bulk protein translation with cycloheximide and assaying substrate levels . Although the knockdown we achieved was quite modest , three of the five substrates were significantly stabilized ( Fig 3B ) . One , RASSF3 , was not stabilized , suggesting either that it is a better βTRCP substrate than the others , or that it is targeted by other ubiquitin ligases . UBE4B is a stable protein . ( Note that we detected UBE4B with a specific antibody against this protein , and did not ectopically express it , so its stability is unlikely to be an artifact . ) It is possible that either only a small pool of the substrate was targeted , or that the outcome of ubiquitination of UBE4B is not proteasomal degradation . Several commonly-used approaches identify ubiquitin ligase substrates as those proteins whose abundance is increased by inhibition of the relevant ligase . One key advantage of ligase trapping is that , in contrast to these techniques , it can identify substrates whose bulk turnover is not affected by inhibition of the ligase . To determine more universally which substrates were quantitatively targeted for degradation by βTRCP , we expressed tagged versions of the substrates , inhibited protein synthesis with cycloheximide , and followed the turnover of the substrate in the absence or presence of MLN4924 ( Table 1 and S6 Fig ) . Of the ten substrates examined , three ( CReP , ZNF395 , and SUN2 ) were unstable proteins that were stabilized by MLN4924 , suggesting that their turnover is mediated by βTRCP alone or in combination with other cullin-RING ligases . ( CReP was previously shown to be an unstable protein [34] , as was SUN2 . ) Four ( ZNF704 , FNIP , RASSF3 and AEBP2 ) were not or only partially stabilized by MLN4924 , suggesting that these might be redundantly targeted by βTRCP and a non-CRL ligase . Three proteins ( HIVEP2 , UBE4B , and TRIM9 ) appeared to be constitutively stable , although we cannot rule out that overexpression or epitope tagging of HIVEP2 and TRIM9 led to an artifactual stabilization . βTRCP could be promoting non-degradative ubiquitination of these substrates , or may only ubiquitinate a specific pool . We were initially concerned that treating cells with MG132 would lead to increased background , or skewing of the results . Therefore , we performed two purifications of the βTRCP ligase trap in the absence of MG132 . This purification generated a list with several of the same substrates , but lacking a subset , especially those shown to be unstable in Figs 3B and S6 ( S7 Fig ) . In addition , all of our validations were performed in the absence of MG132 ( Figs 2 , S4 and S5 ) . We wished to further explore the biological significance of CReP turnover . First , we verified that the ubiquitinated CReP pulled down by the βTRCP ligase trap required SCF activity . Indeed , pre-treatment of cells with MLN4924 eliminated the ubiquitinated CReP ( but not unmodified CReP ) pulled down by the βTRCP ligase trap ( Fig 4A ) . Second , we mutated CReP’s single well-conserved βTRCP-binding consensus , as well as the amino acids immediately downstream , which form a second less-well-conserved consensus . The βTRCP consensus is DpSGX ( 1–4 ) pS [46] , with some substitution of acidic amino acids for phosphorylations tolerated . The sequence we mutated in CReP is DDGFDSDSSLSDSD ( marked in S11 Fig ) . Although this sequence lacks the most-conserved DSG motif , many well-documented βTRCP substrates have variations in this part of the degron [18] , and human CDC25A and CDC25B have well-validated degrons that contain DDG , just like CReP [25] ( shown in Fig 4B ) . This mutant , CReP11A , was significantly stabilized relative to wild type CReP ( Fig 4C and 4D ) , strongly suggesting that CReP turnover is dependent on βTRCP . The notable downshift of the mutant is likely due to mutation of several negatively-charged residues . Mutation of a portion of the same region was independently shown to stabilize CReP while our manuscript was in the review process[47] . Because both protein-folding stress and DNA damage have been shown to regulate eIF2α phosphorylation , we tested whether these stresses also regulated CReP levels . The proteostatic stress inducer thapsigargin had a very minor effect on CReP levels , consistent with a previous report showing no effect [34] . However , DNA damage provoked by either ultraviolet light ( UV ) or the topoisomerase inhibitor camptothecin ( CPT ) led to complete depletion of CReP ( Fig 5A ) . Suggestively , the disappearance of CReP was coincident with the induction of eIF2α phosphorylation by these stressors . The depletion of CReP was not due merely to inhibition of translation by eIF2α phosphorylation , as DNA damage also decreases the half-life of CReP compared to no treatment or treatment with proteostatic stressors in a cycloheximide chase ( S11 Fig ) , and CReP still disappears upon DNA damage in mouse embryonic fibroblasts in which Ser51 of eIF2α has been mutated to alanine ( data not shown ) . CReP turnover and subsequent eIF2α phosphorylation is at least partially dependent on βTRCP , as transfection with shRNA against both paralogs of this ligase delays DNA damage-dependent induction of both CReP turnover and eIF2α phosphorylation ( Fig 5B ) . CReP depletion is fully dependent on CRL-mediated degradation , because treatment of cells with the CRL inhibitor MLN4924 prevents CReP depletion ( Fig 5C ) . The residual CReP turnover seen even in cells treated with βTRCP shRNA may reflect our inability to achieve sufficient knockdown of βTRCP , or additional turnover mediated by another CRL . Cullin-mediated turnover of CReP in response to DNA damage was not restricted to HEK293 cells , since it occurs in both primary human fibroblasts ( Fig 5D ) and immortalized mouse embryonic fibroblasts ( MEFs ) ( S12 Fig ) . The CReP11A mutant was not completely stabilized upon DNA damage ( data not shown ) , possibly because DNA damage promotes βTRCP binding to additional sites on CReP . βTRCP has been shown to interact with non-consensus phosphodegrons in MDM2 , suggesting that it may be difficult to identify degrons by sequence alone[48] . Therefore , we mapped phosphorylated residues on CReP to identify any additional degron sequences ( S9 Fig ) . Notably , most phosphosites were observed both with and without CPT . It is possible that the increase in CReP turnover observed upon DNA damage is not due to increased phosphorylation , but to a change in a targeting factor or localization of CReP . However , phosphosites are still likely to be required for turnover . For clustered phosphosites and phosphosites that were near short acidic stretches , we mutated both the phospho-acceptor and all acidic and potential phospho-acceptors in the region . In addition , we mutated one additional weak βTRCP consensus site that was not covered in the phospho-mapping . We then tested the stability of these mutants , in various combinations , in DNA damage ( data not shown ) . CReP31A ( S10 Fig ) was the least mutated allele that was completely stable upon treatment with DNA damage ( Fig 5E and 5F ) . Importantly , this stabilization was not merely an artifact of high starting levels resulting from prioritized transcription or translation , as CReP31A is stable even upon pre-treatment with camptothecin followed by cycloheximide chase ( Fig 5E ) . Like the 11A mutant , CReP31A migrates much more quickly than the endogenous protein , likely due to mutation of many negatively-charged amino acids . To examine the physiologic role of the turnover of CReP upon DNA damage , we determined whether CReP stabilization had an effect on eIF2α phosphorylation . When CReP turnover was inhibited by knockdown of βTRCP , treatment with MLN4924 , or mutation of CReP , phosphorylation of eIF2α was delayed or inhibited to an equivalent degree ( Fig 5B , 5C and 5F ) . This is not specific to HEK293 cells , as MLN4924 also reduced eIF2α phosphorylation after UV treatment in immortalized mouse embryonic fibroblasts ( MEFs ) ( S12 Fig ) . However , primary human fibroblasts ( Fig 6D ) had constitutively high levels of eIF2α phosphorylation , so the effect of CReP turnover was only subtle . This may reflect a greater need for this pathway in fast-growing cells , or the fact that these primary cells were under constant stress . Upon proteostatic stress , eIF2α phosphorylation promotes the translation of the transcription factor ATF4 [49] . ATF4 activates the expression of the transcription factor CHOP [49] , which in turn promotes the transcription of GADD34 [50] . Like CReP , GADD34 is a PP1 targeting subunit that acts on Ser51 of eIF2α [51 , 52] . These PP1 subunits appear to have a dedicated role in regulating eIF2α , since the lethal phenotype of knockout mice lacking both GADD34 and CReP can be rescued by mutating eIF2α Ser51 [51] . Previous reports suggested that GADD34 is induced at late time points after DNA damage in some cell types [53] . We were especially interested in whether DNA damage promoted the destruction of CReP only to replace it with GADD34 . However , we found that UV treatment did not promote GADD34 protein expression , while ER stress induced by thapsigargin did ( Fig 6A ) . This may reflect a cell-type difference between HEK293 cells and cells previously used to show GADD34 induction . Surprisingly , treating cells with UV and thapsigargin simultaneously blocked the thapsigargin-mediated increase in GADD34 protein levels , suggesting that DNA damage somehow dominantly prevents expression of this protein . Inhibition of GADD34 expression by UV treatment could be rescued by simultaneously treating cells with MLN4924 , suggesting that a CRL is involved in blocking GADD34 accumulation . Finally , we examined whether CReP turnover after DNA damage affected rates of translation . After treatment with DNA damage , translation rate was assayed via the SUnSET method [54] , by adding puromycin to the cells for 10 minutes , then detecting the degree of puromycin incorporation into newly translating polypeptides via western blotting with an anti-puromycin antibody . We found that expression of CReP31A , which led to high CReP levels even after treatment with camptothecin and initial recovery from this damage , accelerated the recovery of translation after DNA damage , doubling the translation rate at 2 hours after CPT washout ( Fig 6B and 6C ) . Notably , this effect was not seen with the unstable , ectopically expressed wildtype CReP , although it was expressed at the same level as CReP31A . This effect reproduced several times , although the exact timing varies , likely due to subtle variations in CReP expression levels during transfection .
We have identified and validated thirteen novel substrates of the well-studied ubiquitin ligase βTRCP via Ubiquitin Ligase Trapping . While we were unable to test two of the twenty-eight candidate substrates identified , 88% of the remaining twenty-six were either known or validated novel substrates . While affinity chromatography is often able to identify ligase substrates , these data suggest that Ligase Trapping provides an unprecedented hit rate , making it an especially efficient way to identify new ubiquitin ligase substrates . Moreover , this technology has allowed us to easily validate substrates even if their bulk stability is not affected by βTRCP ubiquitination . Our results for FBW7 suggest another way in which Ligase Trapping can complement currently available techniques . In a previous study , the Clurman lab pulled out all 26 members of the Mediator complex with FBW7 . They used degron prediction and follow-up experiments to identify MED13 and MED13L as the ubiquitylated Fbw7 substrates and carefully confirmed that they are direct substrates . Our mass spec of the Fbw7 ligase trap immunoprecipitation specifically purified MED13 ( and MED13L ) uniquely in the Fbw7 Ligase Trap , and not in any of the other purifications . Moreover , we pulled out none of the other 25 subunits . This underscores the usefulness of our technique , especially for the great majority of F box proteins for which no degron consensus is known . Thus , even in cases where Ligase Trapping identifies similar numbers of substrates compared to other techniques , it allows one to quickly identify the directly ubiquitylated substrates . In addition to the substrate CReP , which we followed up in detail , turnover of several of the other substrates is likely to be regulated in response to cell cycle position or stress . Sun2 is a transmembrane protein that spans the inner nuclear envelope and has been implicated in the maintenance of nuclear structure and the regulation of DNA damage . Its turnover by βTRCP may regulate these processes , and its removal from the membrane after ubiquitination may also be a regulated step . Strikingly , four of the eleven novel substrates we validated , ZNF395 , HIVEP1/2 , ZNF704 , and AEBP2 , are transcription factors , as are several known βTRCP substrates , such as Nrf2 and ATF4 . We also identified two substrates that are themselves ubiquitin ligases , UBE4B and TRIM9 , which opens up the possibility of complex mutual regulation . While UBE4B ubiquitination depends on the SCF ( data not shown ) , it is not highly ubiquitinated ( Fig 2 ) , and it appears that the majority of the UBE4B population is stable ( Fig 3B ) . RASSF3 is a candidate tumor suppressor protein that activates p53-dependent apoptosis under appropriate conditions , including DNA damage [55] . Its regulation by βTRCP is consistent with the known role of βTRCP in responding to DNA damage , and may help explain the oncogenic effect of βTRCP overexpression [18] ( along with other known tumor suppressor substrates of βTRCP , such as REST[45] ) . RASSF3 appears to have both stable and unstable pools . This may reflect the relatively small pool of cells undergoing stress at any particular time in an untreated culture . Perturbations such as DNA damage might drive RASSF3 turnover . Our previous studies in yeast [13] showed that 56% of newly-identified SCF substrates were strongly stabilized when the F box in question was mutated . 25% showed small or moderate stabilization , but were still unstable in the F box gene mutant . Finally , 19% appeared stable even in wildtype . We find here that 45% of confirmed novel substrates were stabilized by treatment with a pan-CRL inhibitor , 18% showed no stabilization , and 27% were stable in wildtype . Thus , in both cases only half or fewer novel substrates were quantitatively turned over by the single ligase , although this is likely an underestimate overall , since previously characterized substrates may be biased for this category . While some of these effects could be due to the population assay employed , as noted above , substrates such as Cln3 and Gal4 in yeast , as well as PIP box-containing substrates in humans , are targeted in a way that is dependent upon the sub-cellular localization/context of the substrate [12 , 56] . Alternatively , some turnover events occur as part of quality control pathways that only target those proteins that are in some way defective . We have implicated βTRCP in the regulation of translation initiation after DNA damage through its turnover of CReP , and shown that DNA damage-induced phosphorylation of eIF2α , because it uniquely requires the depletion of CReP , occurs via a different mechanism from the other stresses known to promote eIF2α phosphorylation , which all promote kinase activation . Previous work has shown that the phosphorylation of eIF2α after UV treatment depends on the kinase Gcn2 [57 , 58] . We propose that this phosphorylation requires both Gcn2 activation and CReP turnover . Why does phosphorylation of eIF2α require CReP depletion after DNA damage , but not in response to proteostatic stress ? One possibility is that eIF2alpha kinases are less active after DNA damage than after proteostatic stress . We observed that , once CReP levels begin to drop , eIF2α phosphorylation is much higher upon our UV treatment than after proteostatic stress ( Fig 5A ) . This likely reflects both continued CReP activity and the induction of GADD34 upon proteostatic stress . We showed in Fig 6B and 6C that CReP turnover has a significant effect on translation rates after DNA damage , but substantial inhibition of translation happens even in the absence of CReP turnover . Translation rates are highly redundantly regulated , both via control of eIF2α phosphorylation and via regulation of eIF4 . Our results are consistent with a model in which CReP turnover is important to enforce continued low levels of translation at later timepoints . Moreover , the high levels of eIF2α phosphorylation enabled by CReP turnover in response to DNA damage may allow translational reprograming that leads to induction of DNA damage repair proteins , even as global translation is downregulated . Indeed , translation of several DNA repair proteins has been shown to be resistant to inhibition of CAP-dependent translational inhibition by eIF2α phosphorylation [58] . Finally , how do CRLs prevent the induction of GADD34 after UV treatment ? One possibility is that CReP turnover upon DNA damage ( which requires CRLs ) drives such strong eIF2α phosphorylation that translation of GADD34 or one of its upstream regulators ATF4 or CHOP is inhibited . Another possibility is that a CRL is turning over a specific protein to keep GADD34 levels low . βTRCP is known to target ATF4 [24] and the Cul3-associated ligase SPOP is reported to target CHOP [59] . GADD34 is also a known proteasome target , consistent with its being a substrate of βTRCP or another CRL [60] . Targeting of both CReP and Gadd34 for degradation upon DNA damage underscores the importance of limiting eIF2α phosphatase activity during DNA damage .
All plasmids were transfected into the 293 FlpIn TRex cell line ( Life Technologies , Grand Island , NY , USA ) , which contains both a site for FRT-mediated recombination ( which we did not use in this work ) and expresses the tet repressor , which allows doxycycline-inducible expression from promoters that include tet operators . Mouse embryonic fibroblasts ( MEFs ) were immortalized by transduction with the SV40 large T antigen ( kind gift of Morgan Truitt and Davide Ruggero ) . All cells were grown in DMEM with 10% heat-inactivated fetal bovine serum . For large-scale purifications , medium was supplemented with 500 U/mL penicillin and 500 μg/mL streptomycin . 6xHis-ubiquitin was expressed from pTB30 , a modified pcDNA3 . 1 vector with a pCMV/TetO promoter expressing 6xHis-Uba52-IRES-6xHis-RPS27A . The parent of this construct was the kind gift of Zhijian Chen , UT Southwestern . The construct was linearized with Pvu I and transfected into 293 FlpIn TRex cells . Stable transfectants were selected with G418 and a clone was selected that expressed at a high level only upon treatment with doxycycline . To make the ligase trap fusion proteins , F box proteins were fused on the C-terminus to 3xFlag followed by the C terminal half of human RAD23B ( Accession #BC020973 . 2 , amino acids 185–410 ) , encoding two UBA domains . Ligase traps βTRCP2 ( FBXW11; Accession #BC026213 . 1 , pTB53 ) , Fbxo24 ( Accession #NM033506 . 2 , pBEN20 ) , and Fbxo6 ( Accession #NM018438 . 5 , pBEN5 ) were expressed as hygromycin resistance-T2A-ligase trap fusions driven by the mouse PGK1 promoter . Each of these constructs also expresses an shRNA against the relevant F box protein ( to which the fusion protein is resistant ) , driven by the mouse U6 promoter . These cassettes were linearized by digestion with Pac I . Fbw7 ( Accession# NM_033632 . 3 , pTB59 ) Ligase Trap was expressed from a pcDNA3 . 1 vector , under the control of the CMV promoter . The vector was linearized with BglII . All linearized plasmids were transfected into the HisUb cell line and stable transfectants were selected with hygromycin . We selected clonal cell lines that expressed moderate levels of the relevant ligase trap . All substrate proteins were tagged on the N-terminus with the 5xHA epitope , and expressed from the CMV promoter in pcDNA3 . 1 , except SUN2 , AEBP2 , ALDH2 , and RASSF3 , which were tagged on the C-terminus . They were transiently transfected into the relevant cell line using Fugene HD at 3 μL/μg DNA ( Promega Corporation , Madison , WI , USA ) or polyethyleneimine ( at 18 μg/μg DNA ) 24–48 hours before the experiment . βTRCP was knocked down with an shRNA targeting both BTRC and FBXW11 , expressed from the pSUPER-puro-retro vector ( under the H1 promoter ) [61] . MG132 is used at 5 μM . MLN4924 is used at 1 μM . Camptothecin is used at 3 μg/mL , unless otherwise specified . Medium was removed from adherent cells and set aside . Cells were covered in 37°C 1X PBS with 0 . 9 mM CaCl2 and 0 . 5 mM MgCl2 , then exposed to 300 J/m2 UV-C , PBS was aspirated , and medium was replaced . For western blotting , cells were lysed in 1X RIPA buffer with protease and phosphatase inhibitors for 30 minutes on ice , insoluble material was spun out , then protein concentrations were measured with BCA Reagent ( Pierce , Thermo Scientific , Rockford , IL , USA ) and normalized before addition of SDS sample buffer with DTT . For Figs S7 ( except for RASSF3 ) and 5C , cells were lysed directly in SDS sample buffer with DTT or βMe . All gels were Criterion Tris-HCl 4–20% gradients ( cat . #345–0034 , BioRad , Hercules , CA , USA ) , except for the gel for the α-HA blot in Fig 2C , which was a 7 . 5% gel ( BioRad cat . #345–0007 ) . Antibodies used were α-HA 16B12 at 1:1 , 000–1:2 , 000 ( cat . #MMS-101R , Covance , Emeryville , CA , USA ) , α-6xHis at 1:1 , 000–1:2 , 000 , α-ubiquitin P4D1 at 1:100 , α-Flag M2 at 1:2 , 000 ( cat . #F3165 , Sigma , St . Louis , MO , USA ) , α-Cul1 at 1:1 , 000 , α-vinculin at 1:1 , 000–1:5 , 000 , α-βactin at 1:1 , 000–1:10 , 000 ( Sigma cat . #A5441 for Fig 4A , Abcam , Cambridge , UK , cat . #ab8226 for all others ) , α-PPP1R15B ( CReP ) at 1:1 , 000–1:5 , 000 ( cat . #14634-1-AP , Proteintech , Chicago , IL , USA ) , and α-GADD34 ( cat . # 10449-1-AP , Proteintech , Chicago , IL , USA ) . α-phosphoS51-eIF2α ( cat . #9721 ) , α-eIF2α ( cat . #9722 ) , α-phosphoS317Chk1 ( cat . #2344 ) , and α-Chk1 ( cat . #2360 ) antibodies were all from Cell Signaling Technologies , Danvers , MA , USA . The α-puromycin antibody 12D10 was from EMD Millipore ( cat . #MABE343 ) . Western blots in Figs 1 , 2A , 2B and 3A were incubated with secondary antibodies fused to horseradish peroxidase and visualized by treatment with Western Lightning ECL ( Perkin Elmer , Waltham , MA , USA ) . Western blots in Figs 2C , 3B and 4 were incubated with fluorescent secondary antibodies and visualized with an Odyssey scanner ( Licor , Lincoln , NE , USA ) . Unless otherwise noted , stable cell lines expressing Ligase Traps were treated with 5 μM MG132 for 4 hours before collection . We grew 100–200 barely sub-confluent 15 cm dishes for each purification , representing approximately 1–3 x 109 cells . Pellets were lysed in 25 mM Hepes-KOH , pH8 , 150 mM K Oac , 10 mM MgCl2 , 5 mM CaCl2 , 20 mM iodoacetamide , 30 μM MG132 , protease inhibitors , and phosphatase inhibitors by sonication , then treated with DNase ( 660 U/mL ) at 4°C for 30 minutes before addition of Nonidet P-40 to 0 . 1% . Samples were spun to remove insoluble material , then incubated with α-Flag M2 magnetic beads ( Sigma , St . Louis , MO , USA ) at 4°C overnight . Beads were washed 5 times in 1X PBS+0 . 1% Nonidet P-40 , then eluted in this wash buffer+0 . 5 mg/mL 3xFlag peptide . The eluate was denatured by addition of 2X volume Buffer B ( 216 mM NaH2PO4 , 16 mM Tris , 9 . 37 M urea , pHed to 8 ) . The sample was then incubated with NiNTA agarose for 3 hours at room temperature . The beads were washed 3X in Buffer B diluted to 8M urea+10 mM imidazole , then 2X in Buffer B diluted to 1 M urea+10mM imidazole . Samples were eluted in 0 . 5 M urea , 300 mM imidazole , 0 . 1% rapigest ( or Nonidet P-40 if not to be used for mass spectrometry ) , 108 mM NaH2PO4 , 8 mM Tris ( pHed to 8 before adding imidazole ) . The immunopurified protein complexes were mixed in a ratio of 1:1 with digestion buffer ( 100 mM Tris-HCl , pH 8 . 5 , 8M urea ) , reduced , alkylated and digested by sequential addition of lys-C and trypsin proteases as previously described[62 , 63] . For identification of phosphorylation site , proteins were digested directly in the excised gel slice using trypsin[62] . Peptide digests desalted and fractionated online using a 50 μM inner diameter fritted fused silica capillary column with a 5 μM pulled electrospray tip and packed in-house with 15 cm of Luna C18 ( 2 ) 3 μM reversed phase particles . The gradient was delivered by an easy-nLC 1000 ultra high pressure chromatography system ( Thermo Scientific ) . MS/MS spectra were collected on a Q-Exactive mass spectrometer ( Thermo Scientific ) [64 , 65] . Data analysis was performed using the ProLuCID , DTASelect2 , and Ascore algorithms as implemented in the Integrated Proteomics Pipeline—IP2 ( Integrated Proteomics Applications , Inc . , San Diego , CA ) [66–69] . Phosphopeptides were identified using a differential modification search that considered a mass shift of +79 . 9663 on serines , threonines and tyrosines . Protein and peptide identifications were filtered using DTASelect and required at least two unique peptides per protein and a peptide-level false positive rate of less than 5% as estimated by a decoy database strategy[70] . Normalized spectral abundance factor ( NSAF ) values were calculated as described and multiplied by 105 to improve readability [71] . We followed the SUnSET protocol [54] . Puromycin was added to culture medium at a final concentration of 10 μg/mL , incubated for 10 minutes at 37°C and 8% CO2 , then medium was replaced with ice-cold PBS with 5 mM EDTA , and cells were sprayed from the dish on ice , spun down at 4°C and flash-frozen . Samples were normalized by protein concentration , and puromycin incorporation was detected by western blotting with a monoclonal anti-puromycin antibody ( 12D10 ) and quantified by densitometry .
|
Approximately 600 human genes encode enzymes that act as ubiquitin ligases , which facilitate the transfer of the small protein ubiquitin to thousands of substrate proteins; “tagging” with ubiquitin often promotes the degradation of the substrate by the proteasome . In this paper , we adapt a technique called Ligase Trapping for use in mammalian cells . Ligase Trapping is a highly accurate method for determining which substrates are targeted by a ubiquitin ligase . Here we use it to identify new substrates of the human cell cycle regulator βTRCP . Our screen was indeed highly accurate , as we were able to validate 88% of the candidate substrates we identified by mass spectrometry . Some of these new substrates were unstable proteins that were stabilized by inhibition of βTRCP , or of the entire class of ubiquitin ligases of which βTRCP is a part . However , others appear to be stable or redundantly-targeted substrates , which have been more difficult to identify with current techniques . This suggests that Ligase Trapping will be able to reliably identify new substrates of human ubiquitin ligases . Further , one of the new βTRCP substrates , CReP , is specifically depleted upon DNA damage , and depletion of CReP contributes to inactivation of the translational machinery upon DNA damage .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
DNA Damage Regulates Translation through β-TRCP Targeting of CReP
|
Genome-wide association studies have shown that pleiotropy is a common phenomenon that can potentially be exploited for enhanced detection of susceptibility loci . We propose heritability informed power optimization ( HIPO ) for conducting powerful pleiotropic analysis using summary-level association statistics . We find optimal linear combinations of association coefficients across traits that are expected to maximize non-centrality parameter for the underlying test statistics , taking into account estimates of heritability , sample size variations and overlaps across the traits . Simulation studies show that the proposed method has correct type I error , robust to population stratification and leads to desired genome-wide enrichment of association signals . Application of the proposed method to publicly available data for three groups of genetically related traits , lipids ( N = 188 , 577 ) , psychiatric diseases ( Ncase = 33 , 332 , Ncontrol = 27 , 888 ) and social science traits ( N ranging between 161 , 460 to 298 , 420 across individual traits ) increased the number of genome-wide significant loci by 12% , 200% and 50% , respectively , compared to those found by analysis of individual traits . Evidence of replication is present for many of these loci in subsequent larger studies for individual traits . HIPO can potentially be extended to high-dimensional phenotypes as a way of dimension reduction to maximize power for subsequent genetic association testing .
Genome-wide association studies of increasingly large sample sizes are continuing to inform genetic basis of complex diseases . These studies have now led to identification of scores of susceptibility SNPs underlying a vast variety of individual complex traits and diseases [1–3] . Moreover , analyses of heritability and effect-size distributions have shown that each trait is likely to be associated with thousands to tens of thousands of additional susceptibility variants , each of which individually has very small effects , but in combination they can explain substantial fraction of trait variation [4–15] . GWAS of increasing sample sizes as well as re-analysis of current studies with powerful statistical methods are expected to lead to identification of many of these additional variants . An approach to increase the power of existing GWAS is to borrow strength across related traits . Comparisons of GWAS discoveries across traits have clearly shown that pleiotropy is a common phenomenon [3 , 14 , 16–19] . Aggregated analysis of multiple related traits have led to identification of novel SNPs that could not be detected through analysis of individual traits alone [20–23] . Further , analysis of genetic correlation using genome-wide panel of SNPs have identified groups of traits that are likely to share many underlying genetic variants of small effects [10 , 12 , 14 , 24 , 25] . As summary-level association statistics from large GWAS are now increasingly accessible , there is a great opportunity to accelerate discoveries through novel cross-trait analysis of these datasets . A variety of methods have been developed in the past decade to increase power of GWAS analysis by combining information across multiple traits [26–37] . Many of these methods have focused on developing test-statistics that are likely to have optimal power for detecting an individual SNP under certain types of alternatives of its shared effects across multiple traits [26 , 30 , 31 , 35 , 38 , 39] . These approaches do not borrow information across SNPs and may be inefficient for analysis of traits that are likely to have major overlap in their underlying genetic architecture . For the analysis of psychiatric diseases , for example , it has been shown that borrowing pleiotropic information across SNPs can be used to improve power of detection of individual SNP associations and genetic risk prediction [40 , 41] . In this article , we propose a novel method for powerful aggregated association analysis for individual SNPs across groups of multiple , highly related , traits informed by genome-wide estimates of heritability and genetic covariance . We derive optimal test-statistics based on orthogonal linear combinations of association coefficients across traits–the directions are expected to maximize genome-wide averages of the underlying non-centrality parameters in a gradually decreasing order . We exploit recent developments in LD-score regression methodology [14 , 42] for estimation of phenotypic and genotypic correlations for implementation of the method using only summary-level results from GWAS . Our method applies to traits measured on different samples with unknown overlap . We evaluate performance of the proposed method through extensive simulation studies using a novel scheme for directly generating summary-level association statistics for large GWAS for multiple traits with possibly overlapping samples . We use the proposed method to analyze summary-statistics available from consortia of GWAS of lipid traits [43] , psychiatric diseases [20] and social science traits [44] . These applications empirically illustrate that HIPO components can be highly enriched with association signals and can identify novel and replicable associations that are not identifiable at comparable level of significance based on analysis of the individual traits .
Suppose that the summary level results are available for K traits . For a given SNP j , let β^j and sj denote vectors of length K containing estimates of regression parameters and associated standard errors , respectively , for the K traits . Let M be the total number of SNPs under study . Throughout , we will assume both genotypes and phenotypes are standardized to have mean 0 and variance 1 . Let Nk denote the sample size for GWAS for the k-th trait . For binary traits , Nk is the effective sample size Ncase , kNcontrol , kNcase , k+Ncontrol , k . We assume Nk can vary across studies because traits may be measured on distinct , but potentially overlapping , samples ( see Section C in S1 Appendix for discussion of the case where sample size varies across SNPs within the same study ) . We assume that summary-level statistics in GWAS are obtained based on one SNP at a time analysis and that β^j|βj follows a multivariate normal distribution: N ( βj , Σβ^j ) , where βj = ( βj1 , … , βjK ) T is referred to as the “marginal” effect sizes , the coefficients that will be obtained by fitting single-SNP regression models across the individual traits in the underlying population . The variance-covariance matric Σβ^j , which may include non-zero covariance terms when the studies have overlapping samples , will be estimated based on estimates of standard errors of the individual coefficients ( sj ) and estimate of “phenotypic correlation” that could be obtained based on LD-score regression . Power has a one-to-one correspondence with the non-centrality parameter ( NCP , denoted by δ ) of the underlying χ2-statistic . Therefore , we try to find the linear combination cTβ^ that maximizes the average NCP across SNPs ( denoted by E[δ] ) , which is given by E[δ]=E[ ( cTβ ) 2]var ( cTβ^ ) . ( 1 ) Here c is the vector of weights for a HIPO component associated with individual traits; we drop the subscript j from β^j and βj and use β^ and β for simple notations . The denominator is easy to simplify: var ( cTβ^ ) =cTΣβ^c , which does not depend on true value of β . We derive an expression of the numerator based on commonly used random effect models that are used to characterize genetic variance-covariances . Let βj ( J ) = ( βj1 ( J ) , … , βjK ( J ) ) T denote the vector of “joint” effect sizes associated with SNP j that could be obtained by simultaneous analysis of SNPs in multivariate models across the K individual traits . We assume that βj ( J ) follows a multivariate normal distribution N ( 0 , ΣgM ) , where Σg is the genetic covariance matrix . It follows that βj , the vector of marginal regression coefficients , is also normally distributed with mean 0 and E[βjTβj|lj]=ljΣgM , where lj=∑j′=1Mrjj′2 is the LD score . Here rjj′ is the correlation of genotypes between SNP j and j′ . Thus , based on the above model , the numerator of ( 1 ) can be written as E[ ( cTβ]2]=cTE[E[ββT|l]]c=E[l]McTΣgc . Therefore , we have E[δ]=E[l]McTΣgccTΣβ^c . The matrix Σβ^ needs to take into account the sample size differences and overlaps across studies . When all the phenotypes are measured on the same set of people , Σβ^ is proportional to the phenotypic variance-covariance matrix and E[δ] reduces to maximizing the heritability of the combined traits cTy ( MaxH ) [34] . Here y = ( y1 , … , yK ) is the vector of phenotypes . But HIPO is more general and can be applied to traits measured on different samples with unknown overlap . The LD-score regression allows estimation of both Σg and Σβ^ based on underlying slope and intercept parameters , respectively , using GWAS summary-level statistics ( Section A in S1 Appendix ) [14 , 42] . The first HIPO component c1 is given by solving the following optimization problem: maxccTΣg^csubjecttocTΣβ^^c=1 . Subsequent components ck are defined iteratively by solving a slightly different optimization problem maxccTΣg^csubjecttocTΣβ^^c=1andcTΣβ^^cl=0 ( l=1 , 2 , … , k-1 ) . The above procedure can be implemented by suitable eigen decomposition , resulting in a total of K HIPO components ( Section B in S1 Appendix ) . We call the first HIPO component HIPO-D1 , the second HIPO component HIPO-D2 , and so on . Interestingly , it can be shown that the eigenvalues resulting from this procedure are the average NCP for χ2 association-statistics across SNPs along the HIPO directions ( up to the same scale constant , Section B in S1 Appendix ) . Ideally , it is adequate to consider the top HIPO components if a few eigenvalues clearly dominate the others . For the kth HIPO component , the association for the SNP j is tested using Z-statistics in the form zj , ck , =ckTβ^jckTΣβ^^ck . It is easy to see that HIPO z-statistics reduce to the inverse standard error weighted z-scores when all traits have the same heritability , have genetic correlation 1 and , there is no sample overlap across studies . Therefore , HIPO can also be viewed as an extension of standard single-trait meta-analysis . It can be expected from theory that the performance of HIPO , characterized by the increase of average NCP of HIPO components compared to individual traits , does not directly depend on the overlaps of causal SNPs across traits or the overlap of samples across studies . In fact , E[δ] only depends on the covariance matrices var ( β ) and var ( β^ ) . These two matrices can stay the same under different degrees of causal SNP overlap and sample overlap . A closely related method is MTAG , which uses the summary level data of multiple traits to estimate single trait effects , based on genetic and phenotypic correlation across traits [36] . MTAG is also based on linear combinations of summary statistics but the weights are different from those obtained by HIPO . More specifically , MTAG solves the moment equation E[β^j-ωkωkkβj , k]=0 , which gives the solution β^MTAG , j , k=ωkTωkk ( Ω-ωkωkTωkk+Σβ^ ) -1ωkTωkk ( Ω-ωkωkTωkk+Σβ^ ) -1ωkωkkβ^j . Here var ( βj ) = Ω , and ωk is a vector equal to the k-th column of Ω and ωkk is a scalar equal to the k-th diagonal element of Ω . The matrix Σβ^ is the same for all SNPs if both genotypes and phenotypes are standardized to have mean 0 and variance 1 . We will compare HIPO with MTAG in simulations and real data analysis . We use a novel simulation method that directly generates summary level data for GWAS of multiple traits preserving realistic genotypic and phenotypic correlation structures . We proposed the single-trait version of this approach in a recent study [15] . We propose to simulate GWAS estimate for marginal effects across K traits , denoted as β^j= ( β^j1 , … , β^jK ) T , using a model of the form β^j=βj+vj+ej , where two types of errors terms , vj and ej , are introduced to account for variability due to population stratification effects and estimation uncertainty , respectively . We assume the population stratification effects vj s follow independent and identically distributed ( i . i . d . ) multivariate normal across SNPs . We generate the estimation error terms e~= ( e1T , … , eMT ) T following a multivariate normal distribution that takes into account both phenotypic correlation across traits and linkage disequilibrium across SNPs . Note that there is widespread correlation between error terms , which can exist across different SNPs within the same study or for the same SNP across different studies . Correlation can also exist between different SNPs in different studies in the presence of LD , phenotypic correlation and sample overlap . All the possibilities can be captured by simulating from e~~N ( 0 , R⨂Σe ) where the covariance matrix is the Kronecker product of the LD coefficient matrix R={rjj′}j , j′=1 , … , M and Σe={NklNkNlcov ( yk , yl ) }k , l=1 , 2 , …K where the ( k , l ) element involves sample sizes , the sample overlap Nkl and the phenotypic covariance between the kth and lth trait ( Section D in S1 Appendix ) . We assume that the sample size is the same for all the SNPs within the same study . We simulate βj by first randomly selecting ~12K causal SNPs out of a reference panel of ~1 . 2 million HapMap3 SNPs with MAF >5% in 1000 Genomes European population . This SNP list is downloaded from LD Hub [45] . For selected casual SNPs , we generate i . i . d . joint effect sizes βj ( J ) from a multivariate normal distribution N ( 0 , Σg12 , 000 ) , where Σg is the genetic covariance matrix . For simplicity we assume all the traits have the same set of causal SNPs . We calculate the marginal effect sizes βj as the sum of the joint effect size of SNPs in neighborhood Nj weighted by the LD coefficient , i . e . βj=∑j′∈Njβj′ ( J ) rjj′ . The neighborhood Nj is defined to be set of SNPs that are within 1MB distance and have r2 > 0 . 01 with respect to SNP j . For simulation of e~ , we observe that in a GWAS study where the phenotypes have no association with any of the markers , the summary-level association statistics is expected to follow the same multivariate distribution as e~ . We utilize individual level genotype data available from 489 European samples from the 1000 Genomes Project . For each of the 489 subjects , we simulate a vector of phenotype from a predetermined multivariate normal distribution without any reference to their genotypes . We then conducted standard one SNP at a time GWAS analysis for each trait to compute the association statistics β^j , 1000G= ( β^j1 , 1000G , … , β^jK , 1000G ) T for the 1 . 2 million SNPs . To mimic the incomplete sample overlap between traits , we can calculate β^j1 , 1000G , …β^jK , 1000G based on different subsamples of 1000 Genomes EUR , of size n1 , … , nK . Finally , to generate error terms according to sample size specification for our simulation studies , we use the adjustment ej= ( n1N1β^j1 , 1000G , … , nKNKβ^jK , 1000G ) T , We show in Section D in S1 Appendix that this e~= ( e1T , … , eMT ) T has the desired distribution . We conduct simulation studies to validate HIPO-based association tests and investigate expected power gain under varying sample size and heritability . For simplicity , we first consider the scenarios where all traits are measured on the same set of subjects . To make the settings more realistic , we use two sets of genetic and phenotypic covariance matrices estimated from real data: We choose only 3 psychiatric diseases instead of all 5 involved in real data analysis to speed up computation . ASD , BIP and SCZ have high heritability and substantial genetic correlation , which should be helpful to illustrate the property of the method . We vary the value of scale factor hmax2=0 . 1 , 0 . 2 , 0 . 35 , 0 . 5 to control heritability of the traits while preserving the genetic correlation structure . We also vary the sample size: N = 10K , 50K , 100K , 500K . The covariance matrix of vj is set to 7 . 35×10-8 ( 10 . 50 . 50 . 50 . 510 . 50 . 50 . 50 . 510 . 50 . 50 . 50 . 51 ) and7 . 35×10-8 ( 10 . 50 . 50 . 510 . 50 . 50 . 51 ) in the first and second settings , respectively . This choice of parameters leads to an average per SNP population stratification that is about 25% of the per SNP heritability when hmax2=0 . 35 . For each setting we repeat the simulation 100 times . More extensive simulations are conducted to investigate the performance of HIPO under partial sample overlap across studies , and when different traits have different sets of causal SNPs ( S1 Table ) . We also investigate the performance in higher dimensions by simulating 10 traits which are divided into two blocks of 5 traits , with higher correlation within blocks and lower correlation between blocks ( S1 Table ) . In addition , we conduct simulations using UK Biobank data to study the type I error of HIPO under unbalanced case-control design ( Section E . 1 of S1 Appendix ) , as well as the relationship between the number of dominant HIPO components and underlying genetic mechanisms ( Section E . 2 in S1 Appendix ) . We analyze publicly available GWAS summary-level results across three groups of traits measured on European ancestry samples using the proposed method . Global Lipids Genetics Consortium ( GLGC ) provides the GWAS results for levels of low-density lipoprotein ( LDL ) cholesterol , high-density lipoprotein ( HDL ) cholesterol , triglycerides ( TG ) and total cholesterol ( TC ) [43] . The data consist of 188 , 577 European-ancestry individuals with ~1 . 8 million SNPs after implementing the LD Hub quality control procedure ( described at the end of this section ) . The Psychiatric Genomics Consortium ( PGC ) cross-disorder study analyzed data for 5 psychiatric disorders: autism spectrum disorder ( ASD ) , attention deficit-hyperactivity disorder ( ADHD ) , bipolar disorder ( BIP ) , major depressive disorder ( MDD ) and schizophrenia ( SCZ ) [20 , 46–49] . Two of the five traits involved trio data: ASD ( 4788 trio cases , 4788 trio pseudocontrols , 161 cases , 526 controls , equivalent to 4949 cases and 5314 controls ) and ADHD ( 1947 trio cases , 1947 trio pseudocontrols , 840 cases , 688 controls , equivalent to 2787 cases and 2635 controls ) . The other three studies did not involve trios: BIP ( 6990 cases , 4820 controls ) , MDD ( 9227 cases , 7383 controls ) and SCZ ( 9379 cases , 7736 controls ) . After applying the same QC procedure , we included ~1 . 05 million SNPs for HIPO analysis . The Social Science Genetic Association Consortium ( SSGAC ) provides summary statistics for depressive symptoms ( DS , N = 161 , 460 ) , neuroticism ( NEU , N = 170 , 911 ) and subjective well-being ( SWB , N = 298 , 420 ) [44] . The DS data are the meta-analysis results combining a study by the Psychiatric Genomics Consortium [48] , the initial release of UK Biobank ( UKB ) [50] and the Resource for Genetic Epidemiology Research on Aging cohort ( dbGap , phs000674 . v1 . p1 ) . For neuroticism , the study pooled summary level data sets from UKB and Genetics Personality Consortium ( GPC ) . The SWB data is the meta-analysis result from 59 cohorts [44] . All subjects are of European ancestry . We analyzed ~2 . 1 million SNPs after QC . For all three groups of traits , we use the GWAS parameter estimates and standard errors to compute the z-statistics and p-values without making post-meta-analysis correction of genomic control factors . We perform SNP filtering to all three groups of phenotypes based on LD Hub quality control guideline . Markers that meet the following conditions are removed: ( 1 ) with extremely large effect size ( χ2 > 80 ) ( to avoid the results to be unduly influenced by outliers ) ( 2 ) within the major histocompatibility complex ( MHC ) region ( 26Mb~34Mb on chromosome 6 ) ( 3 ) MAF less than 5% in 1000 Genomes Project Phase 3 European samples ( 4 ) sample size less than 0 . 67 times the 90th percentile of the total sample size ( to account for SNP missingness ) ( 5 ) alleles do not match the 1000 Genomes alleles . We further remove SNPs that are missing for at least one trait . The summary statistics are supplied to LDSC software [14 , 42 , 45] to fit LD score regression . We defined a locus to be “novel” if it contains at least one SNP that reach genome-wide significance ( p-value < 5×10−8 ) by the HIPO method and the lead SNP in the region is at least 0 . 5 Mb away and has r2 < 0 . 1 from all lead SNPs of genome-wide significance regions identified by individual trait analysis ( Section F in S1 Appendix ) .
Simulation results show that all HIPO components maintain the correct type I error rate with or without population stratification , consistently across different sample sizes and values of heritability ( Table 1 and S2–S5 Tables ) , even in unbalanced case-control studies ( S13 Table ) . One representative example is the case with covariance structure of blood lipids ( Table 1 ) . It can be seen that correct type I errors are maintained under three different significance levels: p<0 . 05 , p<0 . 01 and p<0 . 001 . Desired type I error rates are also maintained in simulation settings where causal SNPs across traits only overlapped partially and there are incomplete sample overlaps across studies ( S3 Table ) . Similar patterns can be observed for the case with covariance structure of psychiatric diseases ( S4 and S5 Tables ) . In the presence of population stratification , the degree of which is modest according to our simulation scheme , tests based on individual traits show somewhat inflated type I error under large sample size ( e . g . 500K ) ( Table 1 and S4 Table ) . Results also show that association analysis based on HIPO-D1 leads to substantial number of additional true discoveries compared to that based on the most heritable trait ( S6 Table ) . In most settings , the value of average χ2-statistics are larger for HIPO-D1 than those for the individual traits and MTAG estimates ( S7 and S8 Tables ) . After LD-pruning , HIPO components identify a substantial number of novel loci that are not discovered by individual trait analysis ( Table 2 , S9 and S10 Tables ) . For the case with covariance structure of blood lipids , the largest power increase can be observed when N = 50K and N = 100K , where HIPO can identify up to 206 new loci ( Table 2 ) . Similar patterns are followed in other settings ( S9 and S10 Tables ) . Results also show that QQ plots for HIPO-D1 to be more enriched with signals than those for the most heritable trait ( S1–S4 Figs ) . We compare HIPO and MTAG in all our simulation settings . Both MTAG and HIPO find many novel loci compared to individual trait analysis and the set of novel loci identified by HIPO and MTAG tend to have some overlaps , but each method identifies some unique ones that is not discovered by the other . Although MTAG tends to find more loci than HIPO in total , HIPO tends to find more novel loci that are not detected by individual trait analysis and MTAG tends to replicate existing findings ( Table 2 , S9 and S10 Tables ) . This pattern is more obvious when the sample size is large ( N ≥ 100K ) and is not sensitive to the LD clumping criterion ( S11 Table ) . HIPO tends to be more powerful in settings of higher dimensions , with the number of novel loci nearly twice of the number identified by MTAG in several cases ( S10 Table ) . We applied our method to the Global Lipids Genetics Consortium ( GLGC ) data [43] . The average NCP decreases from 0 . 213 for HIPO-D1 to 0 . 026 for HIPO-D4 , with most association signals appears to be associated with the first and second components ( S14 Table ) . This may be due to the fact that LDL and TC are highly correlated with each other and as are HDL and TG , and HIPO-D1 and HIPO-D2 each captures the signal of one pair of traits . HIPO-D1 is positively related to TG , negatively related to HDL and TC and depends weakly on LDL . HIPO-D2 depends mostly on TC . The last component HIPO-D4 , which contains very little genetic association signals and hence can be used as a negative control , is positively correlated with TC and negatively with the other three traits . Note that all 4 traits have similar heritability and sample size ( S14 Table ) , hence the difference in weights is likely driven by genetic correlations . The order of λGC and average of empirical χ2 statistic also tracks with the average NCP ( Fig 1 ) , suggesting that the observed enrichments are likely due to polygenic effects . We identified twenty novel loci by HIPO-D1 and 4 by HIPO-D2 at genome-wide significant level ( p < 5 × 10−8 ) ( Table 3 ) . The number of novel loci changed to 16 for HIPO-D1 and 4 for HIPO-D2 under a more stringent LD-pruning criterion: r2 < 0 . 1 and lead SNPs of different loci are >1Mb from each other . The pattern of p-values for individual traits show that the proposed method detects novel SNPs that contain moderate degree of association signals across multiple traits . There is very little overlap between new loci found by HIPO-D1 and by HIPO-D2 , as expected from genetic orthogonality of the two components ( S5 Fig ) . Among the 24 new loci found by HIPO-D1 and HIPO-D2 , 9 are not found by any of the MTAG estimates . MTAG identified 10 novel loci that are not detected by HIPO-D1 or HIPO-D2 ( S17 Table ) . The result is fairly consistent with our simulation studies which shows that MTAG and HIPO tend to find some overlapping and some unique novel loci . Applications of HIPO to Psychiatric Genomics Consortium ( PGC ) cross-disorder data [20] show that most association signals are captured by HIPO-D1 ( S15 Table ) , which has an average NCP twice larger than that of HIPO-D2 . The first HIPO component puts the highest weights on BIP and SCZ , which have the largest heritability and relatively large sample sizes . It is noteworthy that for a few of the strongest signals , HIPO is outperformed by standard meta-analysis , which was implemented in PGC cross-disorder analysis as a way for detecting SNPs that may be associated with multiple traits . The QQ plot of HIPO-D1 , however , dominates those for the individual traits and for the standard meta-analysis when p > 1 × 10−8 ( Fig 1 ) . This suggests that HIPO is superior to standard meta-analysis in detecting moderate effects , while sacrificing some efficiency for the top hits . The value of λGC and average χ2 -statistics are higher for HIPO-D1 than those for individual traits and standard meta-analysis . HIPO-D1 discovers one new locus , marked by the lead SNP rs13072940 ( p = 1 . 71 × 10−8 ) , that is not identified by either the individual traits or the meta-analysis . The same loci is identified under LD-pruning criterion r2 < 0 . 1 and lead SNPs of different loci are >1Mb from each other . The marker rs13072940 shows association with bipolar disorder ( pBIP = 0 . 0026 ) and schizophrenia ( pSCZ = 2 . 55 × 10−6 ) but no association with autism spectrum disorder ( pASD = 0 . 97 ) , ADHD ( pADHD = 0 . 70 ) or major depressive disorder ( pMDD = 0 . 11 ) . The meta-analysis signal ( pMeta = 7 . 02 × 10−6 ) does not reach genome-wide significance and is , in fact , weaker compared to that from schizophrenia alone . This SNP shows stronger association in more recent larger studies of bipolar disorder [47] ( pBIP = 0 . 0003 ) and schizophrenia [51] ( pSCZ = 1 . 32 × 10−7 ) , clearly indicating that this is likely to be a true signal underlying multiple PGC traits . MTAG also identifies the same new locus ( rs13072940 ) as HIPO-D1 under the same significance and LD pruning threshold . Application of HIPO to Social Science Genetic Association Consortium studies reveals that most of the genetic variation is captured by HIPO-D1 that has an average NCP twice larger than that of HIPO-D2 ( S16 Table ) . The component is negatively associated with DS and NEU and is positively associated with SWB . The tail region of QQ plot of HIPO-D1 lies close to that of neuroticism , but the values of λGC and average χ2 are substantially larger for HIPO-D1 ( Fig 1 ) . HIPO-D1 identifies 12 new loci that are not discovered by individual trait analysis of SSGAC data ( Table 4 ) , increasing the total number of genome-wide significant loci from 24 to 36 ( S7 Fig ) . The number of new loci changes to 11 under a more stringent LD pruning criterion: r2 < 0 . 1 and lead SNPs of different loci are >1Mb from each other . MTAG identifies 14 loci that are not identified by individual trait analysis , including the 12 loci found by HIPO-D1 ( S18 Table ) . We examined evidence of replication of the novel loci based on more recent and larger studies of DS and SWB that were incorporated in the MTAG analysis [36] . As this study reported only a list of top SNPs ( p < 1 × 10−5 ) after stringent LD-pruning ( r2 < 0 . 1 ) , we could not look up the exact lead SNPs that we report for the novel regions ( Table 4 ) . Instead , we searched for SNPs in the top list reported by MTAG study that could be considered proxy ( D’>0 . 75 ) for our lead SNPs . We found 7 of the 12 novel have such proxies and these proxy SNPs show stronger level of association in the more recent MTAG study for at least one of DS and SWB ( Table 5 ) .
In this report , we present a novel method for powerful pleiotropic analysis using summary level data across multiple traits , accounting for both heritability and sample size variations . Application of the proposed method to three groups of genetically related trait identifies a variety of novel and replicable loci that were not detectable by analysis of individual traits at comparable level of confidence . We also conduct extensive simulation studies in realistic settings of large GWAS to demonstrate the ability of the method to maintain type-I error , achieve robustness to population stratification and enhance detection of novel loci . The novel method we introduce for directly simulating summary-level GWAS statistics , preserving expected correlation structure across both traits and SNP markers , will allow rapid evaluation of alternative methods for pleiotropic analysis in settings of large complex GWAS more feasible in the future . Application of the proposed method provides new insight into the genetic architecture of groups of related traits . For blood lipids , which have similar sample sizes , the average NCPs for HIPO-D1 and HIPO-D2 dominate the other two , suggesting that there are perhaps two unrelated mechanisms through which most genetic markers are associated with the individual cholesterol traits . For psychiatric diseases and social science traits , the top HIPO component dominates the others , indicating that there is perhaps one major genetic mechanism underlying each group of traits . These conjectures are supported by a simple simulation ( Section E . 2 , S1 Appendix ) . However , given that top HIPO component down weights traits with smaller sample sizes , it is possible that there exist other independent genetic mechanisms related to these traits that could not be captured by the top HIPO component . Nevertheless , HIPO , by taking into account both heritability and sample sizes , provides a clear guideline how many independent sets of tests should be performed across the different traits to capture most of the genetic signals . Throughout the paper we report results based on significance threshold p < 5 × 10−8 . We do not recommend adjusting the significance threshold to account for multiple comparison , since HIPO components are not independent of individual-trait tests . Similar issues will also arise about MTAG or other pleiotropic methods . Investigation of setting up proper threshold is beyond the scope of the current paper . However , even if we apply Bonferroni correction , HIPO is still able to find a substantial number of new loci . For blood lipids , since there are 4 traits and 2 HIPO components under consideration , the Bonferroni adjusted threshold is p<5×10-86=8 . 3×10-9 . HIPO-D1 still discovers 11 new loci under the new threshold and HIPO-D2 still discovers 1 ( Table 3 ) . For social science traits , since there are 3 traits and 1 HIPO component under consideration , the Bonferroni-corrected threshold is p<5×10-84=1 . 25×10-8 . HIPO-D1 still discovers 7 new loci under the new threshold ( Table 4 ) . Earlier studies have proposed methods for association analysis in GWAS informed by heritability analysis . For analysis of multivariate traits observed on the same set of individuals , the MaxH [34] method was proposed to conduct association analysis along directions that maximizes trait heritability . HIPO allows a generalization of this approach by taking into account sample size differences and overlaps across studies allowing powerful cross-disorder analysis using only summary-level data across distinct studies . Another closely related method is MTAG [36] , which also utilizes summary level data and LD score regression to estimate genotypic and phenotypic variance-covariance matrices . MTAG , however , performs association tests for each individual trait by improving estimation of the underlying association coefficients using cross-trait variance-covariance structure . In contrast , we propose finding optimal linear combination of association coefficients across traits that will maximize the power for detecting underlying common signals . The advantage of MTAG is that it does associate the SNPs to individual traits and thus has appealing interpretation . However , strictly speaking , MTAG , similar to HIPO , is only a valid method for testing the global null hypothesis of no association of a SNP across any of the traits and may identify a SNP to be associated with a null trait while in truth it is only related to another trait in the same group . The advantage of HIPO is that it directly focuses on optimization of power in orthogonal directions for cross-disorder analysis and can provide significant dimension reduction for analysis of higher dimensional traits . Simulation studies as well as analysis of real data shows that both HIPO and MTAG identify substantial number of novel loci compared to analysis of individual traits ( Table 2 , S9 , S10 , S17 and S18 Tables ) . The sets of novel loci identified by the two methods tend to be substantially non-overlapping indicating that it may be useful to implement both methods for cross-trait analysis . Simulations also show that when the number of traits become larger , HIPO tends to find substantially more novel SNPs than MTAG ( S10 Table ) . There exists a variety of methods for pleiotropic analysis [30 , 31 , 35 , 38 , 39] that aim to optimize power for testing associations with respect to individual SNPs without being informed by heritability . The method ASSET [39] , for example , searches through different subsets of traits to find the optimal subset that yields the strongest meta-analysis z statistic for each individual SNP . Methods like HIPO and MTAG , which use estimates of heritability based on genome-wide set of markers , are likely to be more powerful when the underlying traits have strong genetic correlation , such as that observed for psychiatric disorders . In contrast , methods such as ASSET may be more powerful for analysis of groups of traits that have more moderate genetic correlation , such as cancers of different sites [13] , for detection of loci with unique but insightful pleiotropic patterns of association . There is potential to develop intermediate methods , which borrows information across SNPs but in a more localized manner , for example , based on functional annotation information [52 , 53] . Further research is also merited for implementation of HIPO for very high-dimensional pleiotropic analysis and rare variant association studies , two settings in both of which there could be challenges associated with dealing with noises associated with estimation of genetic variance-covariance matrices . In conclusion , HIPO provides a novel and powerful method for joint association analysis across multiple traits using summary-level statistics . Application of the method to multiple datasets shows that it provides unique insight into genetic architecture of groups of related traits and can identify substantial number of novel loci compared to analysis of individual traits . Further extension of the method is merited for facilitating more interpretable and parsimonious association analysis across groups of high-dimensional correlated traits . Our R package is available at https://github . com/gqi/hipo .
|
Pleiotropy is a common phenomenon in genetics that one genetic variant has effects on multiple traits . The shared genetic information across correlated traits can potentially be exploited for enhanced detection of susceptibility loci . Most existing multi-trait methods borrow information across phenotypes but not across SNPs , which can be inefficient for traits that have major overlap . We propose a method that borrows information both across traits and across SNPs to conduct powerful association analysis using summary-level data . Simulations show that the method has correct type-I error rate and substantial increase in power . Application to blood lipids , psychiatric diseases and social science traits identified plenty of new loci that cannot be detected by individual trait analysis . Our method can potentially be extended to high-dimensional phenotypes as a dimension reduction technique .
|
[
"Abstract",
"Introduction",
"Material",
"and",
"methods",
"Results",
"Discussion"
] |
[
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"random",
"variables",
"covariance",
"mathematics",
"genome",
"analysis",
"lipid",
"structure",
"trait",
"locus",
"analysis",
"molecular",
"genetics",
"lipids",
"molecular",
"biology",
"genetic",
"loci",
"probability",
"theory",
"biochemistry",
"blood",
"anatomy",
"heredity",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"computational",
"biology",
"human",
"genetics"
] |
2018
|
Heritability informed power optimization (HIPO) leads to enhanced detection of genetic associations across multiple traits
|
Japanese encephalitis ( JE ) is a mosquito-borne zoonotic disease caused by the Japanese encephalitis virus ( JEV ) . Pigs and water birds are the main amplifying and maintenance hosts of the virus . In this study , we conducted a JEV survey in mosquitoes captured in pig farms and water bird wetland habitats in Taiwan during 2005 to 2012 . A total of 102 , 633 mosquitoes were collected . Culex tritaeniorhynchus was the most common mosquito species found in the pig farms and wetlands . Among the 26 mosquito species collected , 11 tested positive for JEV by RT-PCR , including Cx . tritaeniorhynchus , Cx . annulus , Anopheles sinensis , Armigeres subalbatus , and Cx . fuscocephala . Among those testing positive , Cx . tritaeniorhynchus was the predominant vector species for the transmission of JEV genotypes I and III in Taiwan . The JEV infection rate was significantly higher in the mosquitoes from the pig farms than those from the wetlands . A phylogenetic analysis of the JEV envelope gene sequences isolated from the captured mosquitoes demonstrated that the predominant JEV genotype has shifted from genotype III to genotype I ( GI ) , providing evidence for transmission cycle maintenance and multiple introductions of the GI strains in Taiwan during 2008 to 2012 . This study demonstrates the intense JEV transmission activity in Taiwan , highlights the importance of JE vaccination for controlling the epidemic , and provides valuable information for the assessment of the vaccine's efficacy .
Japanese encephalitis is a vector-borne zoonotic disease transmitted by the bite of a JEV-infected mosquito . Although JE is a vaccine preventable disease , JEV infections are still the leading cause of viral encephalitis in Asia [1] , [2] . It is estimated that 67 , 900 JE cases occur annually in JE endemic countries , with an incidence rate of 1 . 8 cases per 100 , 000 individuals [3] . Symptoms of JE include fever , chills , headache , myalgia , weakness , mental disturbances and neurologic symptoms . The mortality rate can reach as high as 30% , and approximately 30–50% of survivors suffer severe neurological damage [4] , [5] . The JEV belongs to the genus Flavivirus of the family Flaviviridae and is a single-stranded positive-sense RNA virus . The viral genome is approximately 11 kb in length and encodes three structural proteins [capsid ( C ) , premembrane ( prM ) , and envelope ( E ) ] , followed by 7 non-structural proteins ( NS1 , NS2A , NS2B , NS3 , NS4A , NS4B , NS5 ) [6] , [7] . According to phylogenetic analysis of the E protein gene sequences , the JEV strains can be classified into 5 distinct genotypes ( genotypes I–V ) [8] , [9] . Recent studies suggested that the epidemic/dominant genotype of JEV has gradually shifted from genotype III ( GIII ) to genotype I ( GI ) in Southeast and East Asian countries in the last two decades [10]–[14] . Japanese encephalitis is an endemic disease in Taiwan and has been designated as a notifiable infectious disease since 1955 . The highest incidence rate of confirmed JE cases ( 2 . 05 per 100 , 000 ) was recorded in 1967 . After the mass JE vaccination program was implemented in 1968 , the incidence rate of confirmed JE cases declined significantly [15]–[17] . From 1998 to 2011 , the annual number of confirmed JE cases ranged from 13 to 37 . In 2012 , 32 JE cases were confirmed , which is equivalent to an incidence rate of 0 . 13 per 100 , 000 individuals . From 1998 to 2012 , the epidemic peak months were June and July . Confirmed cases occurred sporadically throughout Taiwan . Most individuals diagnosed with JE lived near rice paddy fields or pig farms [18] . We previously reported the molecular epidemiology of JEV in Taiwan [19] . The study demonstrated that all known JEV isolates collected before 2008 belonged to GIII . Genotype I JEV strains that were first found in northern Taiwan in 2008 . In the present study , we monitored the dynamics of the genotype transition and genetic variation of JEV and identified the mosquito species potentially involved in the transmission of JEV in Taiwan during 2005–2012 .
Mosquitoes were collected on pig farms near rice paddy fields in the northern ( Yingge District in New Taipei City and Wujie Township in Yilan County ) , central ( Beitun and Wufeng Districts in Taichung City and Shuilin Township in Yunlin County ) , southern ( Xiaying District in Tainan City , Neimen and Alian Districts in Kaohsiung City , Yanpu and Zhutian Townships in Pingtung County ) , and eastern ( Shoufeng and Guangfu Townships in Hualein County ) regions of Taiwan and from wetland habitats for water-birds in the northern ( Beitou District in Taipei City , and Su'ao Township in Yilan County ) and southern ( Qigu and Anping Districts in Tainan City ) regions from 2005 to 2012 ( Figure 1 ) . The mosquitoes were collected using dry ice traps or sweep nets and were transported either alive or on dry ice to the laboratory . Mosquito collections by sweep nets were conducted only on the same day between 18:30 and 20:30 on the pig farms , while dry ice traps were set up overnight from 17:30 to 7:30 the next morning in the pig farms and wetlands . The predominant mosquito species collected by dry ice trap and sweep net were the same at each collection site . Since dry ice trap method had a long duration of time for mosquito collection , more mosquitoes in numbers and species were captured with this method . The mosquitoes were pooled by species , sex , location , and collection date in groups of 1–50 mosquitoes . Only female mosquitoes were analyzed in this study . The mosquito pools were homogenized in a TissueLyzer ( Qiagen GmbH , Hilden , Germany ) with two cycles at 4°C for 90 sec at a frequency of 30 Hz after adding a 3 mm steel ball to each tube . The pools were then clarified by centrifugation . The supernatants were then sterilized by filtration and removed for RNA extraction and virus isolation . Viral RNA was extracted from the mosquito suspensions using the QIAamp viral RNA mini kit ( Qiagen ) . Three sets of primers , including flavivirus-specific ( FL-F1: 5′-GCCATATGG TACATGTGGCTGGGAGC-3′; FL-R3: 5′-GTKATTCTTGTGTCCCAWCCGGCTGTGTCATC-3′; FL-R4: 5′-GTGATGCGRGTGTCCCAGCCRGCKGTGTCATC-3′ ) , JEV-specific ( JE3F1: 5′-CCCTCAGAACCGTCTCGGAA-3′ and JE3R1: 5′-CTATTCCCAGGTGTCAATATGCTGT-3′ ) and JEV GIII-specific ( 10F: 5′-CTGGGAATGGGCAATCGTG-3′ and 5′-TGTCAATGCTTCCCTTCCC-3′ ) primers , were used for the RT-PCR assay [19] , [20] . Real-time RT-PCR was used to screen for JEV in the mosquito pools as previously described [21] . DNA sequences of positive RT-PCR products were determined . JEV positive samples were then subjected to virus isolation . Cell culture techniques using a mosquito C6/36 cell line or plaque assay using the BHK-21 cell line were used for virus isolation as described previously [22] . Viral RNA was extracted from the JEV-infected culture medium using the QIAamp viral RNA mini kit ( Qiagen ) . The primers used for the amplification and sequencing of the complete open reading frame of JEV are listed in Table 1 . The RT-PCR reaction was carried out using the Superscript III One-Step RT-PCR system with Platinum Taq High Fidelity ( Invitrogen ) . The RT-PCR reaction was performed under the following parameters: 55°C for 30 min; 94°C for 2 min; 40 cycles of 94°C for 15 sec , 50°C for 30 sec , and 68°C for 1 min; and a prolonged elongation at 68°C for 5 min . RT-PCR products were purified using the Qiagen QIA quick Gel Extraction kit ( QIAGEN ) . Nucleotide sequences were determined using the ABI Prism automated DNA sequencing kit and the ABI Prism 3700 DNA sequencer ( Applied Biosystems ) according to the manufacturer's protocols . Overlapping nucleotide sequences were combined and edited using the Lasergene software package ( DNASTAR Inc . , Madison , WI ) . Nucleotide sequences of JEV strains were aligned , edited , and analyzed using ClustalW software . The phylogenetic analysis was performed using MEGA 5 ( http://www . megasoftware . net/ ) [23] . The phylogenetic tree was generated using the maximum likelihood method based on the general time-reversible model . The reliability of the analysis was calculated using 1 , 000 bootstrap replications . The nucleotide sequences of 50 JEV strains isolated from the mosquitoes and a strain isolated from a human were submitted to Genbank with the following accession numbers: KF667277–KF667327 . The maximum likelihood estimates ( MLEs ) of the mosquito JEV infection rate in mosquitoes were calculated using the PooledInfRate software by Biggerstaff ( www . cdc . gov/ncidod/dvbid/westnile/software . htm ) [24] . The Chi-squared test was used for comparison of the mosquito JEV infection rates of different species and sampling sites .
Mosquitoes were captured from 16 localities in Taiwan from 2005 to 2012 ( Figure 1 ) . A total of 102 , 633 mosquitoes belonging to the family Culicidae were collected and analyzed . Table 2 shows a summary of the mosquito species , the MLE of the JEV infection rate per 1 , 000 mosquitoes and the mosquito collection sites ( all localities , pig farms , and wetlands ) . Twenty-six mosquito species from 8 genera of the Culicidae family were identified . The most frequently identified species was Cx . tritaeniorhynchus Giles ( 86 . 90% , n = 89 , 189 ) , followed by Cx . sitiens Wiedemann ( 6 . 13% , n = 6 , 295 ) and Anopheles sinensis Wiedemann ( 2 . 57% , n = 2 , 638 ) . Of the 2 , 848 mosquito pools subjected to real-time RT-PCR for the detection of JEV , 499 were positive . The most frequently identified JEV positive mosquito species was Cx . tritaeniorhynchus ( 468 positive pools ) , followed by Cx . annulus ( 9 positive pools ) and An . sinensis ( 6 positive pools ) ; the MLEs of the JEV infection rates per 1 , 000 individuals in these three species were 5 . 85 , 8 . 99 , and 2 . 3 , respectively . There were significant differences between JEV infection rates in Cx . tritaeniorhynchus and An . sinensis at p<0 . 05 ( P = 0 . 0357 ) , and in Cx . annulus and An . sinensis at p<0 . 01 ( P = 0 . 0044 ) , but was no significant difference in Cx . tritaeniorhynchus and Cx . annulus at p<0 . 05 ( P = 0 . 0979 ) . Culex tritaeniorhynchus and An . sinensis were the most frequently identified mosquito species on the pig farms , whereas Cx . tritaeniorhynchus and Cx . sitiens were the most frequent in the wetlands . The MLE for the mosquito JEV infection rate was significantly higher on the pig farms ( 7 . 5 per 1000 ) than in the wetlands ( 1 . 73 per 1000 ) ( p<0 . 01 ) . Table 3 shows the MLEs of the JEV infection rate per 1 , 000 mosquitoes by month and regions . JEV infected mosquitoes first appeared in early May , peaked in June , and then declined in July . The MLEs for the infection rates for May , June , and July were 4 . 98 , 6 . 72 , and 1 . 46 , respectively . May was the peak month in the southern and eastern regions , whereas June was the peak month in the northern and central regions . Figure 2 shows the mosquito JEV infection rates and the numbers of confirmed human JE cases per month during 2005–2012 . Spring and summer were the epidemic seasons of JE in Taiwan , and June and July were the peak months for human JE cases . Table 4 shows the MLEs of the mosquito JEV infection rates in different geographical locations . In general , the JEV infection rate was highest in the eastern region ( 12 . 32 per 1000 ) ( p<0 . 01 ) , followed by central ( 6 . 02 per 1000 ) and northern ( 6 . 01 per 1000 ) regions , and lowest in the southern region ( 1 . 37 per 1000 ) ( p<0 . 01 ) . Table 4 also shows the numbers of GI and GIII JEV positive pools determined by DNA sequencing of RT-PCR products per year . Of the 499 mosquito pools that were JEV positive by real-time RT-PCR , 374 pools were genotyped by sequencing the real time-RT-PCR products . The real-time RT-PCR was performed using two sets of primers , one primer set targeting a region of the nonstructural protein 5 ( NS5 ) genes to detect all of the flaviviruses , and the other primer set targeting a region of the 3′ untranslated region ( 3′UTR ) to detect JEV . Both partial NS5 gene sequence ( 154 bp ) and 3′UTR sequence ( 220 bp ) contain sufficient information to differentiate genotypes of JEV . All of the JEV positive pools were genotyped each year , except in 2005 and 2009 , only representative RT-PCR positive samples were selected ( based on the place and date of collections ) for sequencing and genotyping . Figure 3A–3E shows the proportional distribution of the GI and GIII JEV strains identified in the northern , central , southern , eastern , and all sites of Taiwan between 2005 and 2012 . The annual numbers of RT-PCR positive pools for genotype analysis were 99 , 42 , 2 , 42 , 47 , 56 , 33 and 53 , respectively , during 2005 to 2012 ( Table 4 ) . Before 2008 , all the JEV found in Taiwan belonged to GIII . GI was first identified in northern Taiwan in 2008 . Since then , the proportion of GI isolates in Taiwan has increased rapidly . From 2009 to 2010 , GI became the predominant JEV genotype circulating in Taiwan . Since 2011 , almost all of the JEV isolates obtained in Taiwan have belonged to GI , with the exception of 2 GIII strains found in Kuantu Nature Park in Taipei City in 2012 . Because GIII was the only JEV genotype identified in 2005–2007 and GI was the only genotype found in 2011 ( Table 5 ) , to estimate the mosquito JEV infection rates according to genotype , we compared the difference between the JEV infection rates in 2005–2007 and 2011 . The GIII JEV infection rate per 1 , 000 mosquitoes in 2005–2007 was 8 . 86 , and the GI JEV infection rate was 6 . 01 in 2011 , there was no significant difference between JEV infection rates in these two groups at p<0 . 01 ( P = 0 . 0294 , chi-squared test ) . Although JEV infection rates for both GIII and GI were not the highest in Cx . tritaeniorhynchus , this mosquito species was the most dominant species harboring the JEV . Among 374 JEV RT-PCR positive pools that were genotyped , 89 . 0% ( 203/228 ) of GIII and 96 . 6% ( 141/146 ) of GI were Cx . tritaeniorhynchus mosquitoes . The GIII JEV infection rate per 1 , 000 mosquitoes of Cx . tritaeniorhynchus was 8 . 90 in 2005–2007 , and the GI rate was 6 . 26 in 2011 , there was no significant difference between JEV infection rates in these two groups ( P = 0 . 0514 , chi-squared test ) . A total of 148 JEV isolates ( 147 from Cx . tritaeniorhynchus and one from Cx . annulus ) were obtained by virus isolation , and the complete E gene sequences of these isolates were determined . In this study , 50 E gene sequences covered the entire sequence diversity of JEV in Taiwan were selected for phylogenetic analysis . The JEV strains isolated from mosquitoes in Taiwan during 2005–2012 fell into two genotypes ( GI and GIII ) . Figure 4A shows the phylogenetic tree of the E gene sequences of GI , which can be grouped into 2 clusters . Cluster 1 contains the JEV strains isolated from mosquitoes collected throughout the country during 2008–2012 , including the first two GI isolates ( TPC0806c and YL0806f ) identified in Taiwan . The GI virus strains in Cluster 2 were first identified in northern and central Taiwan in 2009; in the following year , these strains were found throughout Taiwan . Cluster 2 also contained a JEV strain isolated from a patient ( H10100739 ) who lived in Kaohsiung City . These strains are most closely related to the viruses from China and Japan . These results indicate that multiple introductions and transmission cycle maintenance of the GI strains occurred during 2008–2012 . Figure 4B shows the phylogenetic tree of GIII JEV . The strains isolated in Taiwan between 2005 and 2012 were divided into 2 clusters ( Clusters 1 and 2 ) . The majority of the GIII strains isolated in Taiwan during 2005 to 2011 were classified as Cluster 1 . However , in our study , no strains belonging to this lineage were found in 2012 . Cluster 2 of GIII included a minor group of JEV strains in Taiwan . Most of the isolates were found in the northern and eastern parts of Taiwan . In 2012 , only 2 isolates ( TPC1206c-1 , TPC1206c-2 ) belonging to GIII JEV were found in Taiwan . The Cluster 2 strains of GIII are closely related to viruses from China , Japan , Indonesia and the Philippines .
In this study , we reported the results of a survey of JEV-infected mosquitoes from pig farms and wetlands in Taiwan during 2005 to 2012 . Pig farms near rice paddy fields and wetland habitats for water birds are common in Taiwan , and these places provide suitable environments for the JEV infection cycle [25] . Confirmed JE cases have been identified throughout Taiwan with most of the infected individuals residing near pig farms or rice paddy fields [18] . Cx . tritaeniorhynchus was the predominant mosquito species captured from both the pig farms ( 92 . 6% ) and the wetlands ( 76 . 2% ) and was the main species infected with JEV genotypes I and III . Un-baited sweep net sampling method is considered a passive method that can be used to collect a wide variety of mosquito species . Dry ice ( CO2 ) -baited trap can attract host-seeking female mosquitoes , and CO2 appears to be universally attractive to a variety of mosquito species [26] . Therefore , these two mosquito sampling methods do not seem to have a collection bias towards Cx . tritaeniorhynchus or any of the other mosquito species . These results provide evidence that Cx . tritaeniorhynchus remains a principal vector for the transmission of JEV in Taiwan . Although other Culex mosquitoes , such as Cx . annulus and Cx . fuscocephala Theobald , were also reported as important JEV vectors [27]–[29] , a relatively low number of these mosquitoes were captured , indicating that they play a minor role in the transmission of JEV in Taiwan . The JEV infection rate of mosquitoes captured on the pig farms ( 7 . 50 per 1000 ) was significantly higher than the rate of those captured in the wetlands ( 1 . 73 per 1000 ) ( p<0 . 01 ) , indicating that pigs played an important role in amplifying JEV . In addition , except for A . sinensis , the infection rates of all of mosquito species collected on the pig farms were higher than the rates of those in the wetlands ( Table 2 ) . In our study , JEV positive mosquitoes were captured in only one of the four wetlands , located in Beitou District , Taipei City . This wetland is near human habitats , where both water birds and pigs may serve as reservoirs or amplifying hosts for the JEV . Interestingly , although 11 species of mosquitoes were RT-PCR positive for JEV , the virus was isolated solely from Cx . tritaeniorhynchus and Cx . annulus . Because both blood-fed and unfed mosquitoes were analyzed , and because the RT-PCR results did not allow differentiation between mosquitoes that were actually infected with JEV from those with residual virus in the blood meals , the MLEs of the JEV infection rates in mosquitoes may have been overestimated in this study . According to the Taiwan CDC's surveillance data , the JE epidemic season has occurred annually between May and October , peaking between June and July in recent decades . These results are in accordance with our mosquito surveillance report , where JEV positive mosquito pools appeared in early May , peaked in June , and then disappeared in July . Because most pigs raised for food on pig farms are not immunized with JEV vaccine , only pigs used for breeding are vaccinated in Taiwan , the rapid decline of the JEV infection rates in the mosquitoes captured on the pig farms might be due to an increase in the JEV antibody positive rates of the pigs . In addition , a relatively low JEV positive mosquito rate during this period was observed due to the high mosquito density between July and September [30] in Taiwan . Before 2008 , all the JEV strains identified in Taiwan belonged to GIII . We first found GI JEV in northern Taiwan in 2008 , since then , virus strains of this genotype have rapidly spread throughout the country [19] , [31] . During 2011–2012 , nearly all the JEV strains found in Taiwan belonged to GI . However , although the JEV genotype shifted dramatically from GIII to GI , no obvious change was found in the annual numbers of confirmed JE cases during this period ( Figure 3F ) , suggesting that the JEV vaccination was still effective against newly introduced GI strains in Taiwan . In our study , the JEV E gene sequence phylogenetic analysis provided evidence for multiple introductions and maintenance of the transmission cycles of the GI strains in Taiwan ( Figure 4A ) . Ecological factors , such as climate and landscape , may influence the geographic distribution of the JEV genotypes [29] , [32] . Taiwan is an island located in the Western Pacific off the southeast coast of China , and the Tropic of Cancer passes through the central part of the island . The climate is warm , rain-water is abundant , and many of the wetlands provide suitable habitats for mosquito vectors and water birds . In addition , because pork and rice are the main agricultural products in Taiwan , pig farms and rice paddy fields are very common in the suburban areas of Taiwan . Now that the new GI strains have been introduced to the ideal transmission environment of Taiwan , the new viruses may be able to establish their transmission cycles in new territories . Gao et al . [14] recently reported that the southernmost region of Asia ( Thailand , Vietnam , and Yunnan Province , China ) may have been the source of GI JEV transmission to the Asian continent including Taiwan . The Clusters 1 and 2 of the GI JEV strains in Taiwan belonged to the lineages of the eastern coastal Asian endemic cycle and the central Asian endemic cycle , respectively , suggesting that the GI JEV strains were most likely introduced from China and Japan to Taiwan in recent years . In our study , we found that GI JEV first appeared in northern Taiwan in 2008 . In the following year , GI strains were found in northern and central Taiwan . Subsequently , these viruses spread across Taiwan ( Figure 3A–D ) . The direction of GI JEV transmission in Taiwan seems to be in accordance with the transmission mode proposed by Gao et al . [14] . However , the reasons why the GI strains replaced the GIII strains within such a short period of time , and the ecological and biological factors involved in this event , are still unclear . Further studies are needed to address these questions . This study demonstrated the intense JEV transmission activity in Taiwan and highlights the importance of JE vaccination to control this epidemic . Continuous monitoring of the JEV strain variations and their gene sequence evolution can provide valuable information for the assessment of the vaccine's efficacy .
|
Japanese encephalitis ( JE ) is a vector-borne zoonotic disease transmitted by the bite of a Japanese encephalitis virus ( JEV ) infected mosquito . Japanese encephalitis is an endemic disease in Taiwan . Before 2008 , all known JEV isolates collected in Taiwan belonged to Genotype III of JEV . Genotype I JEV strains were first found in northern Taiwan in 2008 . In this study , we conducted a survey of JEV in mosquitoes during 2005–2012 . A total of 102 , 633 mosquitoes were collected from pig farms and wetlands . Among the 26 mosquito species collected , 11 tested JEV positive by RT-PCR , including Cx . tritaeniorhynchus , Cx . annulus and An . sinensis . Cx . tritaeniorhynchus was the predominant vector species for transmission of JEV in Taiwan . The JEV infection rate of the mosquitoes captured on the pig farms was significantly higher than the rate of those captured in the wetlands , indicating that pigs played an important role in amplifying JEV . A phylogenetic analysis of the envelope gene sequences of JEV isolated from the mosquitoes demonstrated that the predominant JEV genotype shifted from genotype III to genotype I ( GI ) , providing evidence for multiple introductions and transmission cycle maintenance of GI strains in Taiwan during 2008–2012 .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"viral",
"genomics",
"virology",
"emerging",
"viral",
"diseases",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"genetics",
"microbiology",
"genomics",
"microbial",
"genomics"
] |
2014
|
Molecular Epidemiology of Japanese Encephalitis Virus in Mosquitoes in Taiwan during 2005–2012
|
Epigenomic mapping of enhancer-associated chromatin modifications facilitates the genome-wide discovery of tissue-specific enhancers in vivo . However , reliance on single chromatin marks leads to high rates of false-positive predictions . More sophisticated , integrative methods have been described , but commonly suffer from limited accessibility to the resulting predictions and reduced biological interpretability . Here we present the Limb-Enhancer Genie ( LEG ) , a collection of highly accurate , genome-wide predictions of enhancers in the developing limb , available through a user-friendly online interface . We predict limb enhancers using a combination of >50 published limb-specific datasets and clusters of evolutionarily conserved transcription factor binding sites , taking advantage of the patterns observed at previously in vivo validated elements . By combining different statistical models , our approach outperforms current state-of-the-art methods and provides interpretable measures of feature importance . Our results indicate that including a previously unappreciated score that quantifies tissue-specific nuclease accessibility significantly improves prediction performance . We demonstrate the utility of our approach through in vivo validation of newly predicted elements . Moreover , we describe general features that can guide the type of datasets to include when predicting tissue-specific enhancers genome-wide , while providing an accessible resource to the general biological community and facilitating the functional interpretation of genetic studies of limb malformations .
Mammalian body plans are shaped by the precise spatiotemporal execution of transcriptional programs [1] , which have been shown to rely on the coordinated activity of enhancers [2] . Despite the increased availability of epigenomic data , the identification of these short , cis-regulatory DNA sequences in the vast non-coding portions of mammalian genomes has proven to be a difficult task . Indirect measurements suggest that hundreds of thousands of enhancers populate mammalian genomes [3] , but only a few thousand of them have been validated for their activity in vivo [4] . A wide range of experimental and computational approaches have been applied to the prediction of regions showing enhancer activity in vivo , including: 1 ) Evolutionary conservation [5]; 2 ) Chromatin signatures , such as the binding of the co-activator p300 [6] or the acetylation of lysine residue 27 of histone H3 ( H3K27ac ) [7 , 8]; 3 ) Chromatin accessibility to DNase I digestion [9]; 4 ) Genomic sequence signatures , such as the presence of binding sites for relevant transcription factors ( TFs ) [10]; 5 ) Combinations of the former strategies . Despite significant advancements in enhancer identification , through the generation of genome-wide datasets and their integration using supervised [11–15] or unsupervised [16 , 17] models , all available approaches to date suffer from one or more of the following limitations: 1 ) Lack of integration of chromatin and sequence features that are immediately relevant to the tissue ( s ) and the developmental stage ( s ) under consideration; 2 ) Lack of thorough , biological interpretation of the features driving the prediction , which in turn is a key requirement to instruct experiments and more refined models in the future; 3 ) Lack of appropriate negative controls–e . g . the use of random genomic intervals instead of regions showing ( at least partially ) a known signature of enhancers but failing to display tissue-specific activities when tested in vivo; 4 ) Lack of user-friendly access to the de novo predictions , limiting the value of the resulting resources for the community of experimental as well as computational biologists . In this work we integrate multiple machine learning approaches in order to produce robust predictions of enhancers active in the developing limbs of mouse embryos at embryonic day 11 . 5 ( E11 . 5 ) . By focusing on this well-studied developmental system [18 , 19] , we are able to overcome the limitations described above and outperform previously described state-of-the-art approaches [11 , 12] . First , we exclusively considered datasets generated from embryonic limbs ( with one exception , a DNase I hypersensitivity dataset from headless embryos ) at the relevant developmental time points ( E10 . 5 , E11 . 5 and E12 . 5 ) , including the binding profiles for CTCF , the cohesin complex , and a large panel of histone modifications [3 , 6 , 8 , 20 , 21] . Among the latter we also included recently published ChIP-seq data from specific limb compartments [22] . Importantly , we trained statistical models that provide intrinsically interpretable measures of feature importance ( LASSO and random forests ) . This allowed us to identify a previously unreported feature capable of significantly improving predictions , namely limb-specific DNase I enrichment . The predictive power of this feature was confirmed using data from other tissues ( central nervous system and facial prominence ) at the same developmental time point . We additionally trained models based on clusters of evolutionary conserved binding sites for those TFs expressed in the developing limbs , and formally integrated these results with the chromatin features described above . We used a set of >200 validated limb-enhancers and ~2 , 000 negatively tested regions corresponding to either validated elements active in tissues other than limb or that were previously selected based on chromatin or sequence features of active enhancers but failed validation due to absence of reproducible reporter activity [4] . Based on our results , we were able to confirm the in vivo activity of three out of four newly predicted enhancers in the vicinity of the Hand2 gene , an important regulator of limb morphogenesis [23 , 24] . Importantly , our genome-wide predictions can be queried through a user-friendly web interface named LEG ( Limb-Enhancer Genie ) , which is available at http://leg . lbl . gov/ . Since a large fraction of limb developmental enhancers are evolutionarily conserved between human and mouse [6] , the user can also input regions from the human genome . The complete set of predictions along with all the sequencing datasets re-analyzed in this study are available for browsing via a public track hub ( see Methods ) on the UCSC genome browser [25] . By providing the community with significantly improved genome-wide maps of the enhancer landscape underlying limb development , our results will assist the functional interpretation of genetic studies assessing human developmental diseases . Moreover , the analysis of the feature importance in the trained models provides novel generalizable insights into the chromatin signature of developmental enhancers that will help guide the design of predictive models in tissues other than limb .
DNase I accessibility and H3K27ac are routinely used to identify tissue-specific putative enhancers in the human and the mouse genomes [7 , 8 , 26 , 27] . We first aimed to determine the sensitivity and specificity of these marks , using the developing limb as a test case ( Fig 1A ) . Even when used in combination , these marks suffer from both high false positive and false negative rates . More than 50% ( 1 , 094/1 , 967 ) of the limb-negative elements in the VISTA collection [4] overlapped H3K27ac-enriched or DNase I accessible regions ( false positives ) . At the same time , a fraction of enhancers truly active in the limbs at E11 . 5 ( 18/234 ) were still missed by both assays ( false negatives ) . These results prompted us to set up a more integrative approach towards more effective limb enhancer discovery . To this aim , we took advantage of >50 recently published limb-specific , genome-wide datasets ( S2 Table ) from four major categories of chromatin features ( namely DNase I accessibility , six histone modification and co-activator p300 , CpG methylation , and the binding of CTCF/cohesin , Fig 1B ) . We chose to use the limb as a model based on the extensive available chromatin data [3 , 8 , 20 , 21] , which includes robust time series spanning three closely spaced developmental time points ( E10 . 5 , E11 . 5 and E12 . 5 ) and hundreds of in vivo validated elements [4] . These datasets comprise two subregion-specific sets ( representative of two important signaling centers in the developing limb , namely the Apical Ectodermal Ridge , or AER , and the Zone of Polarizing Activity , or ZPA , [22] ) . Importantly , including DNase I digestion patterns from whole embryos whose heads had been removed [28] allowed the estimation of tissue-specific DNase I enrichment scores . In addition to the chromatin state , we also incorporated one class of sequence features , i . e . the predicted clusters of evolutionary conserved binding sites for those TFs expressed in the developing limbs . In order to better understand the relative contribution of chromatin and sequence features , our strategy first considered them separately , and then in combination ( Fig 1C ) . To improve robustness , we partitioned our set of 234 limb enhancers ( positive examples ) and 1 , 967 regions negative for activity in the limbs ( 1 , 025 negatively tested regions and 942 showing activity in another tissue , Fig 1A and S1 Table ) into ten equally sized bins with constant ratio of positive to negative observations ( where one bin was used in turn as a test set , whereas the remaining nine constituted the training set ) . The model parameters were learnt using 10-fold cross-validation over the training set . The models trained include least absolute shrinkage and selection operator ( LASSO , [29] ) , support vector machines ( SVM , [30] ) and random forests ( RF , [31] ) . Model performances were compared across the ten independent test sets . Predictions from different models and distinct sets of features were then combined using ridge regression or a weighted sum of ranks approach ( SOR ) . For the final prediction of enhancers genome-wide , the training step was re-iterated using the entire dataset , and the resulting models and their combinations were used to call enhancer regions using a sliding window ( see Methods ) . LASSO and RF allow evaluation of the importance of each feature for model performance . This enabled us to gain insights into the biological relevance of the most predictive features . A notable novelty in our method is the use of a previously unappreciated feature , a score measuring increased tissue-specific DNase I accessibility ( DNase I enrichment ) , into our predictive models . Unexpectedly , we found the headless embryo DNase I accessibility pattern to be well correlated with the DNase I specific for fore- and hindlimbs ( r = 0 . 87 and 0 . 85 , respectively , within the VISTA-dataset ) . This prompted us to hypothesize that including the ratio between the DNase I signal from limbs and the headless embryos would better capture the limb-specific changes in DNase I accessibility , as compared to the limb DNase I signal alone . Unsupervised clustering of the training set followed by visual inspection of the results provided additional evidence to support this hypothesis and revealed further interesting groups ( S1 Fig ) . Next to clusters of negative elements showing predominantly low values across all features , the set contained groups of negatively tested elements showing known chromatin signatures associated with regulatory elements other than enhancers . One group contained mostly promoter-like elements ( high H3K4me3 and H3K9ac ) , while others resembled insulators ( high CTCF and Smc1a ) or Polycomb-associated heterochromatin ( high H3K27me3 ) [32 , 33] . These qualitative observations prompted us to include these chromatin features , along with the DNase I enrichment score as described above , into our machine learning strategy . We first built models of increasing complexity , starting from p300 alone and incrementally adding H3K27ac , DNase I , DNase I enrichment and all the remaining chromatin features ( Fig 2A and 2B ) . The median AUROC ( Area Under the Receiver Operating characteristic Curve ) as well as the median AUPRC ( Area Under the Precision Recall Curve ) steadily increased by including more features ( as assessed on the independent test sets ) . This was observed consistently across the different models . Considering the one showing the highest performances ( namely the radial SVM ) , the median AUPRC when training only on p300 was 0 . 372 , a figure that increased to 0 . 412 when including H3K27ac , to 0 . 502 if considering also DNase I , and finally to 0 . 542 and 0 . 545 if adding the DNase I enrichment or all the features , respectively ( Fig 2B , S3 and S5 Tables ) . Interestingly , models trained only on H3K27ac/p300 and DNase I ( including the enrichment over the head-less embryo samples ) reached a performance almost as high as the full set of chromatin features on the VISTA dataset . However , these additional features are well-known marks for categories of regulatory elements–e . g . insulators and promoters–that are under-represented in our training set but are widespread genome-wide . In fact , by overlapping the CTCF-bound sites in the developing limbs ( which are enriched for insulator sequences ) [21] with the de novo enhancer predictions genome-wide using different feature subsets , the effect of including the additional features became more evident . While models trained on H3K27ac/p300 and DNase I alone showed 10 to 20% overlap with CTCF-bound sites across a wide range of predicted values , models trained on the complete set of features showed less than 5% ( S2 Fig ) , potentially removing a fraction of false positive predictions . A qualitative evaluation of complex loci ( Fig 2C ) indicated that validated limb enhancers show a higher DNase I enrichment ( DNase I versus Headless embryo ) as compared to nearby regions that tested negative in vivo . Given these observations , we then sought to systematically and quantitatively assess the relative importance of the different chromatin features . Specifically , the estimated coefficients from the LASSO and the mean decrease in accuracy estimated by the RF were evaluated . We also estimated a selection probability for each predictor by Bootstrap LASSO ( see Methods ) . The results are summarized in Fig 3 and S9 Table . DNase I accessibility as well as the limb-specific DNase I enrichment were systematically co-selected ( >0 . 98 selection probability by Bootstrap LASSO ) and showed the highest coefficients as well as importance in the RF . Co-selection of these two features further supports a substantial rather than incremental role of tissue-specific DNase I enrichment in identifying active enhancers in vivo . Interestingly , while both the DNase I enrichment values from hind- and forelimbs were often selected and assigned positive coefficients , the forelimb DNase I signal was only selected in 2 out of 1 , 000 bootstrap samples , in contrast to the the hindlimb DNase I signal which was selected every time . We re-ran the Bootstrap LASSO after removing the hindlimb DNase I and found that the DNase I from forelimbs was selected with a probability of one and almost identical performances . This indicates that the dataset from hindlimb might be favored for technical rather than biological reasons . Other features that were selected with high probabilities , but less often than DNase I associated features , were p300 , CTCF , H3K27ac , H3K27me3 and H3K4me3 ( all showing a selection probability >0 . 6 ) . The small size of the training set combined with the multiplicity of classes of regulatory elements , each represented only by a few examples ( S1 Fig ) , was likely responsible for the lower selection probabilities of these features . As expected , p300 and bulk H3K27ac are the most important predictors after DNA accessibility ( selection probability >0 . 75 ) . On the other hand , when we assessed the contribution of the H3K27ac datasets specific for different sub regions of the limb , we found that the ZPA-specific profile was very unlikely to be selected , in contrast to the AER-specific one , which was often included in the LASSO models with a positive coefficient . This region-specific feature is selected >75% of the time , often together with H3K27ac profiles from whole limbs . Thus , our findings highlight the importance of gathering chromatin information at a finer scale in order to be able to identify enhancers with more sub-regional-specific activity . The histone modification H3K4me3 ( and to a lesser extent H3K9ac , which are both usually found at promoter regions [33] ) , was assigned a negative coefficient , as was the mark H3K27me3 , which has been associated with inactive , poised enhancers [34] or more generally with Polycomb-associated heterochromatin . The two proteins CTCF and Smc1a , while often co-bound to DNA on the same genomic elements ( mainly insulators or promoters ) [32] , are assigned coefficients of opposite sign ( negative for CTCF , positive for Smc1a ) , indicating that cohesin but not CTCF is more generally associated with enhancer function . Smc1a was assigned a high importance for the prediction of limb-enhancers by the RF , while it was selected only in 35% of the bootstrap-samples by LASSO . CpG methylation was also found to be rather important in the RF predictions , but very unlikely to be included in the LASSO models . A possible explanation could be the implicit accounting for feature interactions in the RF , which remain unappreciated by the LASSO . Nevertheless , the small size of the training set impinged our ability to explicitly tease out these combinatorial relationships . Taken together , these results are in line with the expectation that different models are able to capture distinct aspects of the data . This prompted us to combine the results from the multiple models into a single , unified predictive score . Two different approaches were applied to this end: ridge regression ( i . e . finding optimal weights to combine the outputs from the single classifiers ) and an approach based on the weighted sum of the individual output-ranks . These strategies led to a significant improvement in the AUPRC as compared to RF ( p-value = 0 . 03 , one-tailed Wilcoxon signed-rank test , Fig 4B ) and a smaller improvement when considering LASSO or linear SVM ( p = 0 . 05 , combined ridge model ) , but it was not significant as compared to radial SVM ( p = 0 . 57 , combined ridge model ) . We then asked whether considering clusters of evolutionary conserved TF-binding sites could lead to a more consistent increase in the predictive power of the combined models . In order to limit the number of input features , we only considered binding motifs for TFs expressed in the developing limbs ( see Methods , S3 and S4 Tables ) . The overall performances of the sequence features alone were markedly lower than those achieved by chromatin ( Fig 4A and 4B , S3 , S4 and S5 Tables ) . Even the application of gkm-SVM [35 , 36] , a motif-agnostic machine learning approach which has been previously applied to enhancer prediction [37] , lead to performances comparable to those achieved by our models ( median AUPRC of 0 . 197 , S11 Table and Discussion ) . However , combining our sequence- and chromatin-based predictions using ridge regression significantly outperformed the combined chromatin model ( Figs 4B , S3 and S4; p = 0 . 019 in terms of AUPRC , one-tailed Wilcoxon signed-rank test; p = 0 . 08 when considering the combined SOR ) and the best single chromatin model ( namely the radial SVM , p = 0 . 024 , one-tailed Wilcoxon signed-rank test ) . Model performances across the test sets are reported in S5 Table . Similar to the chromatin features , we combined the importance measurements from the LASSO and the RF , and short-listed the most relevant TFs ( Fig 4C and S10 Table ) . Interestingly , these TFs are over-represented in publications in the field of limb development as compared to TFs whose motifs were selected less frequently or not selected at all ( Fig 4D , p-value = 0 . 02 , Mann-Whitney test , Top vs No ) . Among the most frequently selected motifs are those of Hoxa13 and Hoxd9 , which are known regulators of digits and stylopod development , respectively [38] . Tp63 is a critical factor for epithelial development that , when mutated , can lead to severe developmental defects including complete absence of limbs [39] . Tfap2a has been previously associated with distal outgrowth of the developing limbs [40] . While Tp63 and Tfap2a have been associated to limb development , these results suggest they might exert their function by binding to enhancer elements . These findings underscore the importance of applying interpretable machine learning approaches to highlight relevant features , in turn helping to formulate new experimental hypotheses . By combining the predictions from both the chromatin- and sequence-based models , our strategy outperformed the state-of-the-art approaches [11 , 12] both in terms of AUPRC and AUROC ( Table 1 ) . In line with this , the distributions of predicted values between the positive and negative regions in the training set showed a stronger separation in our combined models as compared to EnhancerFinder and EMERGE ( Kolmogorov–Smirnov statistic , S5 Fig ) . We also compared the performance of our combined models to the predictive power of the strong enhancers chromatin states defined using two ChromHMM models [17] , trained using eight histone modifications from two distinct biological replicates [41] from E11 . 5 limbs ( see Extended Methods ) . This resulted in a recall of 0 . 162 and a precision of 0 . 447 for the enhancer calls from one of the two replicates . At the same level of recall , our combined models reached a much higher precision ( 0 . 787 and 0 . 741 for the ridge regression and the SOR approach , respectively ) . The second replicate led to comparable conclusions ( recall of 0 . 256 and precision of 0 . 417 for the ChromHMM calls , as compared to a precision of 0 . 732 and 0 . 710 , given that level of recall in our combined models ) . These results prompted us to train the models using the complete set of observations and to run them genome-wide . The mouse genome was tiled into overlapping windows of 2kb , which were assigned prediction values for tissue-specific enhancer activity in vivo using all models . The resulting predictions for each single model ( either LASSO , SVM or RF ) and type of feature considered ( chromatin , sequence ) as well as their combination were ranked , and the 20 , 000 highest scoring regions were binned into 10 groups ( see Methods and S8 , S13 and S14 Tables ) . These were used to evaluate the enrichment for proximity to genes involved in limb development and expressed in Theiler stages 19 and 20 ( corresponding to the window from E11 to E13 ) using GREAT [42] ( Figs 5A and S6 ) . The enrichments from the combined predictions were higher than those of any single model trained on chromatin or clusters of conserved TF-binding sites alone , indicating that the combined models can identify thousands of bona fide , previously uncharacterized , enhancers . Interestingly , while showing lower performances on the test sets as compared to the combined ridge classifier ( Fig 4B ) , the SOR showed the highest enrichment in terms of proximity to genes relevant to limb development ( Fig 5A ) , especially for the highest-ranking elements ( S6 Fig ) . To further corroborate our predictions , we searched the literature for developmental limb enhancers that were robustly validated in vivo but are not part of the VISTA collection . We identified five elements , all of which overlapped the 10 , 000 highest scoring predictions ( considering either the ridge regression or the sum of ranks , S12 Table ) . One of these regions is the ZRS ( ZPA Regulatory Sequence ) , which is a well-known enhancer controlling the expression of Shh in the ZPA [43] . This element consistently ranked very high across both the sequence- and chromatin-based predictions , in line with the abundance of conserved TF-binding sites and the presence of a strong limb-specific DNase I signal ( Fig 5B , left panel , and S12 Table ) . A further , independently identified element that drives the expression of Tfap2a in limbs and face [44] was correctly predicted within an intron of the Tfap2a gene itself ( Fig 5B , right panel ) . We then additionally verified the ability of the proposed approach to identify bona fide limb-enhancers by choosing four newly predicted elements close to the developmental regulator TF Hand2 and testing them in vivo through mouse transgenic enhancer-LacZ reporter assays ( Fig 5C ) . Recently published promoter-Capture-C data [45] from developing limbs at E11 . 5 revealed that these elements are indeed located in a domain contacting the Hand2 promoter with high frequency ( 2/4 reaching statistical significance , S7 Fig ) demonstrating their potential to act as enhancers for this gene . Hand2 displays critical developmental functions in various embryonic tissues such as the limb [23 , 24] , the heart [46–48] and the craniofacial structures [49] . However , the Hand2 limb-specific enhancer landscape has been poorly characterized so far . Three out of four tested elements displayed reproducible LacZ reporter staining at E11 . 5 , with patterns of activity specific to limbs and overlapping well-known subdomains of Hand2 expression [50] . Interestingly , the only element that tested negative also showed the lowest predicted combined score ( Fig 5C and S15 Table ) . Finally , we made the genome-wide predictions available at http://leg . lbl . gov/ . These can be directly and systematically queried through a user-friendly interface . The website also provides two tutorials that leverage published datasets that were not used in the predictions .
In this work , we integrated >50 genome-wide chromatin datasets with sequence information and were able to improve our ability to recognize limb enhancers over previously published approaches . These include EnhancerFinder [11] and EMERGE [12] which represent computational state-of-the-art tools in the field ( Table 1 ) and to our knowledge are the only two studies that employed the VISTA dataset in a way that is comparable to our approach . Combined with making the predictions readily available via a user-friendly interface , another advantage of the presented study is the extensive application of machine learning models ( LASSO , RF ) that are intrinsically designed to provide feature importance . In this way , we have shown that including a limb-specific DNase I enrichment score dramatically improves the prediction of developmental limb-enhancers in terms of both precision and recall over incorporating just the commonly used histone-mark H3K27ac ( Fig 2 ) . The availability of both DNase I accessibility and H3K27ac ChIP-seq data for midbrain , hindbrain , neural tube and facial prominence tissues at E11 . 5 ( S17 Table ) allowed us to reproduce this finding also in tissues other than limb at the same developmental time point . To this aim , we fit logistic regression models including p300 , H3K27ac , DNase I and DNase I enrichment features consecutively , and found that inclusion of the DNase I enrichment scores on top of the other features considerably and significantly improved the performance of all the neuronal tissues and to a lesser extent of facial prominence ( S8 Fig ) . Our ability to evaluate the performances for craniofacial enhancers is affected by the lower number of validated examples ( 74 versus 274 , 310 and 196 for hindbrain , midbrain and neural tube , respectively ) leading to greater variance in the overall performance . In addition to the DNase I enrichment score , the chromatin features H3K4me3 , H3K9ac , H3K27me3 as well as the binding of CTCF and Smc1a were selected as informative ( Fig 3 ) . The inclusion of features other than H3K27ac and DNase I accessibility reflects the presence of VISTA elements showing combinations of these genomic features in the developing limbs , including many that failed to validate in vivo . These are very likely to be insulators , unannotated canonical promoters or poised enhancers , rather than active enhancers ( S1 Fig ) . As many regions in VISTA were selected specifically because they showed canonical enhancer marks , these other classes are likely under-represented in our training data . Nevertheless , these types of elements could be misclassified as enhancers when only DNase I and H3K27ac are used for genome-wide scanning , and our approach proved effective at exploiting this information . When moving to the scale of genome-wide prediction , the small but significant improvements in performance observed when including all these features lead to important differences in the types of elements that are predicted . For example , models trained on H3K27ac and DNase I only are more enriched for insulator-like elements , as indicated by a larger overlap with CTCF-bound regions ( S2 Fig ) . Of note , LASSO also systematically selected with a positive coefficient the H3K27ac dataset specific for the AER sub region ( Fig 3 ) . This demonstrates the value of sub-regional-specific datasets . Nevertheless , there are two issues that need to be addressed in order to fully harness this kind of datasets in the future . First , the current number of tested VISTA enhancers ( and more in general in the literature ) showing activity for each different sub-region is still low . Second of all , there are very few high-quality genomic datasets generated from sub-regional dissected tissues available at present . The performances of the sequence-based models were on the other hand lower than what we observed when incorporating experimentally derived , chromatin data ( Fig 4 ) . The performance of our models are seemingly lower than previously reported [37 , 51] . Previous publications mainly focused on the prediction of sequences with enhancer activity using randomly selected genomic regions ( matched by GC- and repeat- content ) as negative examples . In this study , we focused on a different problem , i . e . the identification of enhancers showing limb-specific activity , against regions that either show enhancer activity in a different tissue than limb , or anyway were selected based on partial experimental evidence of activity , but failed in vivo validation . In fact , when we applied gkm-SVM [35 , 36] , a machine learning approach previously applied to enhancer prediction [37] , we observed performances comparable to those achieved by our sequence-based models ( S11 Table ) . More in general , the identification of transcription factor binding sites genome-wide suffers from a high false positive rate , a problem that was only partially mitigated by leveraging the information of TF-binding-sites-clustering and the evolutionary conservation of these sites . On top of this , developing tissues are complex , heterogeneous mixtures of lineages giving rise to multiple cell-types , each one of which depends on only partially overlapping gene regulatory networks . As such , the diversity of regulatory elements at the sequence level is expected to be much greater in tissues than in more homogeneous , in vitro cell populations . This factor is likely to have a major impact on the signal-to-noise ratio for the identification of sequence-encoded features . In line with this , the most important TFs identified by the sequence-based models are enriched for general regulators involved in enhancer function in the limb , like the Hox family genes , Tp63 or Tfap2a ( Fig 4C and 4D ) . Despite these limitations , we found multiple evidences supporting the value of integrating both chromatin and sequence features in our predictive framework . These included functional analysis of the de novo predictions using GREAT [42] ( Figs 5A and S4 ) , our in vivo validation of three out of four newly predicted enhancers very likely involved in the transcriptional regulation of Hand2 ( Fig 5C and S15 Table ) , and the recapture of previously validated limb-enhancers from a number of independent studies ( Fig 5B and S12 Table ) . Of note , even though the incorporation of the sequence features significantly improved the predictions per se ( Fig 4A and 4B ) , the use of evolutionary conserved TF-binding sites still led to a considerable number of false positive ( S9 Fig ) . A more unbiased approach to mitigate all the issues highlighted in this paragraph would be the generation of high-quality ChIP-seq profiles for the cell-type-specific TFs involved in the development of the embryonic tissue under study . Overall , our results will help instruct future strategies for the identification of enhancers . Our analysis strongly suggests that the use of a limited number of features relevant to the developing organ system under scrutiny ( chromatin accessibility , high enrichment for H3K27ac and p300 binding and low to no enrichment for H3K27me3 , H3K4me3 and CTCF , see Fig 3 ) , as well as the integration of a previously unappreciated feature , the DNase I enrichment , will likely improve the prediction of enhancers active across development and showing diverse tissue and sub-regional specificity . We expect this to be the case in the near future , as soon as the relevant genome-wide datasets are generated . We envision that measuring relative chromatin accessibility across tissues by means of ATAC-seq [52] might provide the same information ( and in turn the same boost in predicting bona fide enhancers ) as the DNase I enrichment score proposed here . At the same time , while more sophisticated computational models could be applied [53] , these are currently limited by the size of the training set . Data gathering remains the major limiting step ( e . g . validation in transgenic mouse lines is still relatively low-throughput ) . Technological advancements to increase the throughput as well as to standardize the assays ( e . g . by site-specific integration of the reporter transgene in the genome ) will soon be required and extremely beneficial . Importantly , by providing the community with an easy access to significantly improved genome-wide prediction maps of the enhancers active during limb development , we anticipate these results to be of value for both developmental biologists and human geneticists . Our web-interface ( http://leg . lbl . gov/ ) can be queried using human genomic regions . This will specifically help the functional contextualization of human non-coding variants , pinpointing their contribution to limb malformations . As an example , the LEG predictions overlapping published H3K27ac-enriched regions in embryonic human limbs [20] ( S16 Table and Extended Methods ) are provided .
All animal work was reviewed and approved by the Lawrence Berkeley National Laboratory Animal Welfare Committee . All mice used in this study were housed at The Animal Care Facility ( ACF ) at LBNL . Mice were monitored daily for food and water intake , and inspected weekly by the Chair of the Animal Welfare and Research Committee and the head of the animal facility in consultation with the veterinary staff . The LBNL ACF is accredited by the American Association for the Accreditation of Laboratory Animal Care International ( AAALAC , IACUC-approved animal protocol #290008 ) . Human and murine validated elements were downloaded from the VISTA enhancer browser ( http://enhancer . lbl . gov ) [4] and mapped to mm10 coordinates using liftOver [25] . After filtering ( see Extended Methods ) , 2 , 201 elements were used for machine learning ( S1 Table ) . ChIP-seq , DNase-seq , RNA-seq and CpG-methylation profiles collected for this study are listed in S2 Table . ChIP-seq and DNase I hypersensitivity reads were aligned to the mm10 release of the mouse genome ( Dec . 2011 , GRCm38 ) using bowtie2 [54] . ChIP-seq peaks were called using MACS v1 . 4 . 2 [55] for analysis regarding overlaps to enriched regions ( not machine learning ) . RNA-seq datasets were aligned to the reference transcriptome ( mm10 , Ensembl 81 gene annotation release , [56] ) using STAR v2 . 4 . 2a [57] . Transcripts were quantified with Stringtie v1 . 0 . 4 [58] . CpG-methylation bigWig tracks at base-pair resolution were downloaded from the ENCODE repository ( http://www . encodeproject . org/ ) [3] . Log2-RPKM quantifications for ChIP-seq and DNase I samples for each one of the 2 , 201 mm10-mapped VISTA elements were performed after expanding them to a minimum size of 2kb around their center . For ChIP-seq samples , enrichments were computed relative to the corresponding control samples ( input DNA ) ( see Extended Methods ) . Scaling of the input features was performed as z-scores . For CpG-methylation , the average fraction of methylated CpGs was determined for each region . Position weight matrices ( PWMs ) [59] ( S3 and S4 Tables ) were limited to those representing binding preferences of TFs potentially expressed in the developing limb ( see Extended Methods ) . Putative TF-binding sites were identified using FIMO v4 . 10 . 2 [60] , with a p-value cutoff of 10−4 and using GC-content matched backgrounds ( see Extended Methods ) . Clusters were identified using a sliding window ( 500bp ) ; binding sites were weighted by evolutionary sequence conservation , as estimated by phastcons [61] . Either the mouse or the human sequence was scanned according to which version was tested in vivo ( S1 Table ) . A complete table of the scores for each TF-gene across the 2 , 201 VISTA elements is provided in S7 Table . The observations were split into ten equally sized groups . Each group was used as test set exactly once while the rest was used for training . Parameters were tuned by ten-fold cross-validation within each training set ( see Extended Methods ) . For the chromatin data , four different classifiers were trained: 1 ) LASSO logistic regression [29]; 2 ) Support Vector Machines ( SVM ) [30] with linear kernel; 3 ) SVM with radial kernel and 4 ) Random Forests ( RF ) [31] . For the sequence data , radial SVMs were not fit . Fig 1C summarizes the modeling strategy . In order to combine the predictions from the separate models , two methods were applied ( see Extended Methods for a detailed description ) : 1 ) ridge regression ( i . e . finding optimal weights for the output from the single classifiers , or “model stacking” ) and 2 ) the weighted sum of the individual output ranks . Predicted values for each one of the input observations are reported in S6 Table , overall performances in S5 Table . For genome-wide predictions , models were fit using 10-fold cross-validation on the entire dataset . After extensive pre-processing of the values of the single features ( see Extended Methods ) , the mouse genome was tiled into gap-less , overlapping 2kb tiles ( with a step of 1kb ) . Tiles overlapping either gene promoters or elements in the training set were discarded . The top 20 , 000 disjoint elements predicted by each model ( or combination ) were obtained using an iterative merging strategy ( see Extended Methods ) . For the bootstrap LASSO , 1 , 000 bootstrap samples of the original data were extracted ( see Extended Methods ) . Model parameters were estimated and selection probabilities for each feature were calculated by dividing the number of non-zero coefficients across bootstrap samples by the total number of bootstrap samples [62] . For the RF , the importance for each variable was defined as the average decrease in accuracy . The Limb-Enhancer Genie ( LEG ) is an online tool ( available at http://leg . lbl . gov/ ) aimed at facilitating the access to the genome-wide predictions generated in this study . Two separate analysis modes are available . The first one finds the overlap of a set of input regions with the top 10 , 000 predicted limb-enhancers . This can be used , for example , to scan large regions for potential limb-enhancers . The second one is conceived to assign scores to smaller regions ( < = 10kb ) . For each input region , the highest scoring overlapping 2kb genomic tile is identified and returned along with its score and original coordinates . This also allows scoring of elements overlapping the training data or regions close to promoters , which were excluded from the top 10 , 000 reported predictions . This second mode of analysis accepts mouse ( mm9 and mm10 ) as well as human ( hg19 or hg38 ) regions . All the predictions along with tracks for the chromatin features and evolutionary conserved TF-binding sites ( for the TF-features most correlated with activity in limb ) are available on the UCSC Genome Browser [25] ( for both mm10 and mm9 ) via the track hub available at http://portal . nersc . gov/dna/RD/ChIP-Seq/LEG_trackhub/hub . txt . The source code for training and combining the models is available for download at http://github . com/rmonti/limb_enhancer_genie/ . All the described data processing steps were performed in the statistical computing environment R v . 3 . 2 . 1 ( www . r-project . org ) . An overview of the packages used in this study along with references to them is given in the Extended Methods . Newly tested elements were named according to the nomenclature current in use in the VISTA Enhancer Browser ( http://enhancer . lbl . gov/; mm: mouse , hs: human ) . The elements were amplified from mouse genomic DNA and cloned into an hsp68-lacZ expression vector , as previously described [5] . Genomic coordinates are listed in S15 Table . Transgenic mouse assays were conducted as previously described [5 , 63] . Sample sizes were selected empirically based on past experience of performing transgenic mouse assays for >2 , 000 total putative enhancers [4] . Mouse embryos were excluded from further analysis if they did not encode the reporter transgene or if the developmental stage was not correct . All transgenic mice were treated with identical experimental conditions . Randomization and experimenter blinding were unnecessary and not performed . Transgenic mouse assays were performed in Mus musculus FVB strain mice . The E11 . 5 developmental stage was considered . Animals of both sexes were used in the analysis . See the previous paragraph for details on sample size selection and randomization strategies .
|
The majority of the human genome does not code for proteins . Regulatory roles have been ascribed to a growing fraction of the non-coding genome . Enhancers , short stretches of non-coding DNA , confer spatial and temporal specificity to gene expression patterns . These regions are essential to the proper development of multi-cellular organisms and , when mutated , can give rise to congenital malformations and contribute to human disease . In line with these observations , a large fraction of the genetic variation associated to developmental abnormalities cannot be narrowed down to protein-coding defects , suggesting these variants actually reside in functional non-coding regions , such as enhancers . However , identification of enhancers in the mammalian genomes that are functional in vivo remains a difficult task . Here we combine chromatin and DNA-sequence data from the mouse genome in a machine learning framework resulting in the Limb-Enhancer Genie ( LEG ) , an accurate and easily accessible collection of predicted enhancers active in the developing limbs . LEG outperforms state-of-the-art approaches , as testified by the high fraction of newly tested elements that we validated in the developing mouse embryo in vivo . To grant the community access to our predictions and their mappings to the human genome , we established a user-friendly web-interface .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"legs",
"nucleases",
"deoxyribonucleases",
"enzymes",
"gene",
"regulation",
"dna-binding",
"proteins",
"enzymology",
"limbs",
"(anatomy)",
"developmental",
"biology",
"artificial",
"intelligence",
"epigenetics",
"mammalian",
"genomics",
"embryos",
"chromatin",
"embryology",
"computer",
"and",
"information",
"sciences",
"musculoskeletal",
"system",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"animal",
"genomics",
"biochemistry",
"hydrolases",
"enhancer",
"elements",
"cell",
"biology",
"anatomy",
"computer",
"architecture",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"machine",
"learning",
"user",
"interfaces"
] |
2017
|
Limb-Enhancer Genie: An accessible resource of accurate enhancer predictions in the developing limb
|
Expansion of the neocortex is a hallmark of human evolution . However , determining which adaptive mechanisms facilitated its expansion remains an open question . Here we show , using the gyrencephaly index ( GI ) and other physiological and life-history data for 102 mammalian species , that gyrencephaly is an ancestral mammalian trait . We find that variation in GI does not evolve linearly across species , but that mammals constitute two principal groups above and below a GI threshold value of 1 . 5 , approximately equal to 109 neurons , which may be characterized by distinct constellations of physiological and life-history traits . By integrating data on neurogenic period , neuroepithelial founder pool size , cell-cycle length , progenitor-type abundances , and cortical neuron number into discrete mathematical models , we identify symmetric proliferative divisions of basal progenitors in the subventricular zone of the developing neocortex as evolutionarily necessary for generating a 14-fold increase in daily prenatal neuron production , traversal of the GI threshold , and thus establishment of two principal groups . We conclude that , despite considerable neuroanatomical differences , changes in the length of the neurogenic period alone , rather than any novel neurogenic progenitor lineage , are sufficient to explain differences in neuron number and neocortical size between species within the same principal group .
Development of the mammalian , and in particular human , neocortex involves various types of neural stem and progenitor cells that reside in the germinal layers of the cortical wall [1]–[5] . An increase in the proliferative capacity of these cells underlies the evolutionary expansion of the neocortex , notably the increase in neuron number . At the onset of mammalian cortical neurogenesis , neuroepithelial cells transform into apical radial glia ( aRG ) , which repeatedly undergo mitosis at the apical surface of the ventricular zone ( VZ ) and typically divide asymmetrically to self-renew and generate either a neuron , an apical intermediate progenitor cell , a basal intermediate progenitor cell ( bIP ) , or basal radial glia ( bRG ) ( the latter two being collectively referred to as basal progenitors [BPs] ) [1]–[5] . In contrast to aRG cells , BPs delaminate from the apical surface and translocate their nucleus to the basal-most region of the VZ to form a secondary germinal layer , the subventricular zone ( SVZ ) , where they divide symmetrically or asymmetrically . In developing mouse neocortex , bIPs typically divide symmetrically to generate two post-mitotic neurons ( neurogenic bIP ) [5]–[8] , whereas in the macaque and human , bIPs can also frequently undergo symmetric proliferative divisions ( proliferative bIP ) [5] , [9] , [10] . Similarly to aRG cells in the VZ , bRG cells in the SVZ divide both symmetrically and asymmetrically [9] , [11]–[14] , which leads to the proliferation of their population and their self-renewal , respectively [5] . Importantly , the symmetric proliferative divisions of bIPs and bRG cells result in the transit-amplification of BPs [9] , [10] , [12] , [15] , [16] , which in turn allows for an increase in the efficiency of subsequent neuron generation [2] , [5] , [17] , [18] . In mammals exhibiting an abundance of BPs during cortical neurogenesis , the SVZ becomes further compartmentalized into an inner ( ISVZ ) and outer SVZ ( OSVZ ) , as first described in the macaque [19] and subsequently observed in several species in which bRG cells constitute a relatively high proportion of BPs [1]–[3] , [5] . Moreover , bRG cells are characterized by radial fibers , which distinguish them from bIPs . These radial fibers of bRG cells in the OSVZ of gyrencephalic mammals typically have divergent , rather than parallel , trajectories to the cortical plate , which is thought to contribute to creating the folded cortical pattern observed in these species through the tangential expansion of migrating neurons [2] , [3] , [5] , [20] . For this reason , and based on supporting evidence obtained in the gyrencephalic human and ferret and lissencephalic mouse , an abundance of asymmetrically dividing bRG cells in the OSVZ has been thought to be necessary for establishing a relatively large and gyrencephalic neocortex [1] , [9] , [11] , [12] . However , subsequent work in the lissencephalic marmoset ( Callithrix jacchus ) has shown that bRG cells may , in fact , exist in comparable abundance in the developing neocortex of both gyrencephalic and lissencephalic species [21] , [22] , indicating that bRG abundance alone cannot be sufficient for either establishing or increasing cortical gyrification . Rather , the mode of cell division , that is , symmetric proliferative versus asymmetric self-renewing , of bRG cells , and of BPs in general , may be a critical determinant of the extent to which gyrification occurs and the neocortex expands . Notably , despite considerable progress in the study of brain size evolution [23]–[25] , the adaptive mechanism that has evolved along certain mammalian lineages to produce a large and folded neocortex is not known . In this study , we analyzed physiological and life-history data from 102 mammalian species ( Tables S1 and S2; Database S1 ) . We show that a gyrencephalic neocortex is ancestral to all mammals and that GI , like brain size , has increased and decreased along many mammalian lineages . These changes may be reliably characterized by convergent adaptations into two distinct physiological and life-history programs , resulting in a bimodal distribution of mammalian species with regard to the gyrencephaly index ( GI ) and the amount of brain weight produced per gestation day . We explain the appearance of these two groups in mammalian evolution by the adaptation of differences in the lineages and modes of cell division of progenitor cells during corticogenesis . We predict that symmetric proliferative BP divisions are key to evolutionary changes in gyrification and expansion of the neocortex .
We collected GI data ( Figure S1 ) for 102 species sampled from every mammalian order and tested multiple models for GI evolution using a species-level supertree [26] . The model that conferred the most power to explain GI values across the phylogeny while making the fewest assumptions about the data ( i . e . , that had the lowest Akaike Information Criterion [AIC] ) diverged significantly from a null model of stochastic evolution [27] and showed a disproportionate amount of evolutionary change to have occurred recently , rather than ancestrally , in mammals ( Figure S2 ) . We identified a folded neocortex ( GI = 1 . 36±0 . 16 standard error of the mean [SEM] ) as an ancestral mammalian trait ( Figure 1 ) . This result held even when additional hypothetical lissencephalic species were added to the root of the phylogenetic tree ( Table S3 ) . It is apparent from ancestral and other internal node reconstructions ( Figure S3 ) not only that GI is very variable , but also that reductions in the rate at which GI evolves have favored branches leading to decreases in GI ( e . g . , strepsirrhines and eulipotyphla ) and accelerations in that rate have favored branches leading to increases in GI ( e . g . , carnivores and caviomorphs ) . A simulation of the average number of total evolutionary transitions between GI values evidences more affinity for transitioning from high-to-low than low-to-high GI values: the majority of high-to-low transitions ( 58 . 3% ) occurred in species with a GI<1 . 47; and the fewest transitions ( 16 . 7% ) occurred across a threshold GI value ( see below ) of ∼1 . 5 ( Figure S4 ) . This finding indicates that , although there is an evident trend in mammalian history to become increasingly gyrencephalic , the most variability in GI evolution has been concentrated among species below a certain threshold value ( GI = 1 . 5 ) . We therefore present a picture of early mammalian history , contrary to most previous work , but which is gathering evidence through novel approaches [28] , [29] , that the Jurassic-era mammalian ancestor may , indeed , have been a relatively large-brained ( >10 g ) species with a folded neocortex . The evolutionary effects of a folded neocortex on the behavior and biology of a species is not immediately clear . We therefore analyzed associations , across the phylogeny , of GI with discrete character states of 37 physiological and life-history traits ( Table S2 ) . Distinct sets of small but significant ( R2≤0 . 23 , p<0 . 03 ) associations were found for species above and below a GI value of 1 . 5 , indicating that these two groups of species adapt to their environments differently ( Figure 2A ) . Although species above and below GI = 1 . 5 tend to fall within classical definitions of slow and fast life-histories , respectively , our results argue in favor of a dichotomy rather than a continuum and , additionally , bear out ecological and behavioral associations not historically bracketed in slow or fast life-history paradigms [30] . For example , the result that narrow habitat breadth and large population group size are associated with low-GI ( <1 . 5 ) species , whereas wide habitat breadth and large social group size are associated with high-GI ( >1 . 5 ) species , suggests not only that an ecological distinction be made for mammals between the population size of co-habitating individuals and the number of those individuals interacting socially , but also that the number of habitat types in which a species must compete may assert a positive selection pressure on neocortical evolution . Importantly , both the low-GI and high-GI groups are sampled from across the phylogeny , testifying to the absence of a phylogenetic signal in the establishment of the two groups and a functional role for GI in the evolution of life-history programs . Hierarchical clustering analysis also supports a bimodal distribution above and below a GI value of 1 . 5 ( Figures 2B and S5 ) . In order to test the bimodal distribution explicitly , we regressed GI values against neuroanatomical traits typically identified with ( and studied in the field of ) neocortical evolution: brain weight , neocortical volume , and neuron number . We found that each scaling relationship could be explained comparably well by either a non-linear function ( Figure 3A ) or two grade-shifted linear functions , with the best-fit linear models drawing significantly different slopes for high-GI and low-GI species ( Figure 3B–3D ) . Specifically , by plotting GI as a function of cortical neuron number , we were able to determine , with two significantly different linear regressions for high- and low-GI species ( T = 4 . 611 , degree of freedom [d . f . ] = 29 , p = 2 . 8×10−4 ) , demarcating values of 1±0 . 11×109 neurons and 1 . 56±0 . 06 GI ( Figure 3D ) , thus providing a neuron number correlate for the GI threshold . The deviation of these results from previous work , which have shown strong phylogenetic signals associated with both GI [31] , [32] and neuron counts [33] , may be explained both by our more than 2-fold increase in sampled species and the a priori assumption of previous work that GI and neuron number evolve as a function of phylogeny . Variation in GI , therefore , has not evolved linearly across the phylogeny , but has in fact been differentially evolved in two phenotypic groups . By identifying an evolutionary threshold in the degree of gyrencephaly , as well as a correlate in terms of neuron number , we revealed the existence of two neocortical phenotypic groups , which found support in their distinct life-history associations ( i . e . , the GI is bimodally distributed and supports two principal mammalian phenotypes ) . These groups could be further divorced by accounting for the amount of brain weight accumulated per gestation day—a confident proxy for neonate brain weight per neurogenic period ( Figure S6A and 6B ) —which we show to be , on average , 14-times greater in high- compared to low-GI species ( Figure 4 ) . Notably , each GI group is constituted by both altricial and precocial species , so the degree of pre- versus post-natal development is not enough to explain the discrepancy in brain weight per gestation day in each group . Rather , to explain the discrepancy , we introduced a deterministic model of cortical neurogenesis , using series summarizing seven neurogenic lineages ( Figure 5A and 5B ) and based on cell-cycle length , neuroepithelial founder pool size , neurogenic period , and estimates of relative progenitor-type population sizes ( Tables 1 and 2 ) . In total , 17 species were incorporated in the model , as we were limited by the number of species for which cortical neuron number was available . These species include species from four phylogenetic orders: Primata , Scandentia , Rodentia , and Didelphimorphia . We arrived at two models that show the highest reliability for predicting cortical neuron numbers in a range of species: a mouse neurogenic program , which implicates only asymmetrically dividing aRG and bRG cells and terminally dividing IPs ( Figure 5A , lineages 1–3 ) ; and a human neurogenic program , which additionally implicates BPs undergoing symmetric proliferative divisions in the SVZ ( Figure 5A , lineages 4–7 ) . Each model is defined by the proportional occurrence of each lineage in that model ( Table 2 ) . Using the mouse neurogenic program we were able to predict neuron counts within 2% of the observed counts for mouse and rat , but underestimated neuron counts by more than 80% in high-GI species ( Figure 5C; Table S4 ) . Increased proportional occurrences of the bRG lineage 3 ( Figure 5A ) with increasing brain size was required to achieve estimates with <5% deviation from observed neuron counts in the other low-GI species ( Table 2; Figure S7 ) . The human neurogenic program predicted neuron counts within 5% for all six high-GI species , but overestimated neuron counts by more than 150% for the low-GI species . Estimates of proportional occurrences of the various lineages in the mouse , marmoset , rabbit , macaque , and human are supported by previous work detailing relative abundances of different progenitor cell-types during cortical neurogenesis [7] , [9]–[11] , [14] , [22] ( IK and WBH , in preparation ) . Evolutionary gain or loss of proliferative potential in the SVZ is an essential mechanistic determinant of neocortical expansion , such that the presence of symmetric proliferative BP divisions in high-GI species and their absence in low-GI species is sufficient and even requisite for explaining neocortical evolution ( Figure S8 ) . Notably , the lissencephalic opossum , a marsupial species with extreme altriciality , required a decreased proportional occurrence of the bIP-containing lineage 2 ( Figure 5A ) and an increased proportional occurrence of the direct neurogenic lineage 1 ( Figure 5A ) but , like all species analyzed here , could not achieve its observed neuron count without the bRG-containing lineage 3 ( Figure 5A ) . This suggests that bRG cells are ancestral at least to the therian stem [34] . To simulate the adaptiveness of evolving increased proliferative potential in the SVZ in two lissencephalic species—mouse and marmoset—we calculated trade-offs between neuroepithelial founder pool size and neurogenic period using mouse/marmoset and human programs of cortical neurogenesis to achieve 109 neurons . We show that , in both species , evolving a lineage of BPs capable of symmetric proliferative divisions is between two and six times more cost-efficient than either expanding founder pool size or lengthening neurogenesis; and that the marmoset , by evolving such proliferative BPs , could achieve 109 neurons by increasing either its observed founder pool or neurogenic period less than 15% ( Figure 6 ) . We further clarified the significance to neuron output of each progenitor-type with deterministic and stochastic models of temporal dynamics and progenitor cell-type variables . The proportional contributions of each lineage to overall neuron output in the mouse and human neurogenic programs were calculated using stage-structure Lefkovitch matrices . By excluding lineages one at a time , we determined the degree to which each lineage contributed to total neuron production . From these analyses , it was clear that symmetric proliferative BPs are increasingly necessary in larger brains and that any exponential increase in neuron production is statistically implausible in the absence of such BPs ( Table S5 ) . Finally , we described the dynamics of asymmetric self-renewing versus symmetric proliferative progenitors , isolated from their observed lineage beginning at the apical ( ventricular ) surface , by introducing three ordinary differential equations ( ODEs ) modeling a self-renewing cell that generates either a differentiated cell or proliferative cell . The ODEs describe a self-renewing mother progenitor , which can generate either a post-mitotic neuron or a proliferative daughter at each division . The proliferative daughter is allowed one symmetric proliferative division followed by self-consumption . The likelihood of a neuron or proliferative daughter being generated by the mother , therefore , is interdependent . We also include the pool of mother progenitors as a linear variable . We show that neuron output of the system increases dramatically when both the initial pool of self-renewing cells and the likelihood of those initial cells to generate proliferative , rather than differentiated , cells approaches saturation ( Figure S9 ) .
We have shown that a gyrencephalic neocortex is ancestral to mammals . This finding is concordant with evidence [29] that the mammalian ancestor was relatively large ( >1 kg ) and long-lived ( >25-year lifespan ) and , furthermore , provides considerable resolution to recent evidence for a gyrencephalic eutherian ancestor [28] by sampling nearly twice as many species and categorizing gyrencephaly as a continuous , rather than a binary , trait . More surprisingly , we show that convergent evolution of higher orders of gyrencephaly along divergent lineages has been accompanied by two distinct constellations of physiological and life-history paradigms . Specifically , species with a GI>1 . 5 , which is commensurate with 1 billion cortical neurons , exhibit patterns of development and life-history that are distinct from species with a GI≤1 . 5 , irrespective of phylogeny . This implies that there is a considerable constraint on either the ability of species of a given neocortical size to exploit certain ecologies or the potential for species of a given ecology to freely adapt neocortical size . Even marine mammals , whose selection pressures are sui generis , may largely be held to the same evolutionary stereotyping as terrestrial mammals ( Figure S10 ) . While our results countenance previous studies showing associations between physiological and life-history traits in mammals ( see [40] ) , we identify those traits to have a bimodal distribution , rather than to vary allometrically , across species . This distribution depicts a Waddington-type landscape for neocortical expansion—albeit relevant at the species-level—wherein the GI threshold represents an adaptive peak requiring a particular adaptation in neurogenic programming within a population for traversal . Our results may explain this landscape by mechanistic differences occurring during cortical neurogenesis between species above and below the GI threshold: the necessity of symmetric proliferative BPs in high-GI species and their putative absence in low-GI species . The human neurogenic program constructed here clearly shows that the same neurogenic lineages in the same proportions are required to generate the neocortices of Old World monkeys , apes , and humans , and may even be extended to carnivores , cetartiodactlys , and other high-GI species ( Figure S10 ) , demonstrating that neurogenic period alone may be sufficient to explain differences in neocortical size between any species in the same GI group ( Figure S11 ) . Our data are insufficient , however , to determine whether these adult differences are uniform across the neocortex or differentially represented in infra- versus supra-granular layers [20] , [41] . We propose that symmetric proliferative divisions of BPs , in addition to having an abundance of bRGs in an expanded SVZ , are necessary and sufficient for the evolution of an expanded and highly folded neocortex in mammals . Recent work in the fetal macaque supports this proposal [10] . We thus conclude that an increase in the proliferative potential of BPs is an adaptive requirement for traversing the evolutionary GI threshold identified here . But because we reconstruct the eutherian ancestor to have a GI value of 1 . 48±0 . 13 ( standard error of the mean [SEM] ) ( Table S3 ) , which falls within the range of the observed threshold , we are left with an ambivalent evolutionary history for mammalian neocortical expansion: either ( i ) BPs capable of undergoing symmetric proliferative divisions are ancestral to all eutherian mammals and were selected against along multiple lineages ( e . g . , rodents , strepsirrhines ) , so that the ultimate loss of BP proliferative potential in certain taxa , and therefore the evolution of low-GI species , is the result of divergent developmental adaptations; or ( ii ) such symmetric proliferative BPs are not ancestral to eutherian mammals , but evolved convergently along multiple lineages , in which case the developmental process for their inclusion in neurogenic programming may be conserved , even if that process was unexpressed for long stretches of mammalian evolution . We have revealed an important insight into mammalian evolution: a GI threshold exists in mammalian brain evolution; neocortical expansion beyond that threshold requires a specific class of progenitor cell-type ( BPs ) to adopt a specific mode of cell division ( symmetric proliferative ) ; and the difference in neuron output between any species on the same side of that threshold does not appear to require adaptations to the lineage or mode of cell division during neurogenesis , but may simply reflect differences in the length of the neurogenic period . Further research into the conservation of genomic regions regulating the capacity of BPs to undergo symmetric proliferative divisions ( e . g . , through the establishment and maintenance of a proliferative niche in the SVZ ) in low- versus high-GI species may reveal whether this mechanism for neocortical expansion has evolved independently in distantly related species or is the product of a deep homology in mammalian cortical development .
We calculated GI using images of Nissl-stained coronal sections from http://brainmuseum . org . We used 10–22 sections , equally spaced along the anterior-posterior axis of the brain , for each species ( Figure S1 ) . The inner and outer contours of the left hemisphere were traced in Fiji ( http://fiji . sc/wiki/index . php/Fiji ) . The species for which we calculated GI are indicated by an asterisk in Table S1 . Additional GI values were collected from the literature ( Table S1; Database S1 ) . Several species ( e . g . , platypus ) , whose cortical folding has been described [42] , [43] but not measured according to the method established by [44] , could not be included in our primary reconstructions of GI evolution ( Figure 1 ) . However , these species , assumed to be lissencephalic ( GI = 1 . 0 ) , were included in supplemental analyses ( Table S3; see Reconstructing the evolutionary history of GI ) . Work in humans and baboons has shown that interindividual variation in GI is not enough to outweigh interspecific differences [45] , [46] . Variation in the mode and tempo of a continuous character trait is not always best characterized by a random walk ( i . e . , Brownian motion ) . Therefore , we compared a range of evolutionary models on the phylogenetic distribution of GI to find the best fit for the data [47]–[50] . Log-likelihood scores for each model were tried against the random walk score using the cumulative distribution function of the χ2 distribution . Maximum-likelihood ancestral character states of GI and rate-shifts in the evolution of GI were then constructed using the best-fit model , with the standard error and confidence intervals calculated from root node reconstruction in PDAP using independent contrasts [51]–[53] . Although a number of putatively lissencephalic non-eutherians were unavailable for our analyses ( see Calculating GI ) , we nonetheless reconstructed alternative ancestral GI values that included one hypothetical monotreme and three hypothetical marsupials ( Table S3 ) . The phylogeny used in this analysis was derived from a species-level supertree [26] . We appreciate that the phylogenetic hypothesis reconstructed by [54] gives notably deeper divergence dates for mammalian sub-classes; however , not enough of our sampled species were included in this reconstruction for it to be useful here . To trace evolutionary changes in GI at individual nodes and along lineages , we used a two-rate mode that highlighted the differences in high ( >1 ) versus low ( <1 ) root-to-tip substitutions and then sampled rates based on posterior probabilities across the tree using a Monte Carlo Markov Chain . We assumed that transitioning between adjacent GI values had the highest likelihood of occurrence . The rate at a given node could then be compared to the rate at the subsequent node to determine if a rate transition was likely . We corroborated these results using the auteur package [55] , which calculates rate-transitions at internal nodes under the assumption of an Ornstein-Uhlenbeck selection model [34] over 1 million Monte Carlo sampling iterations drawn from random samplings of posterior distributions of lineage-specific rates . Scaling relationships were determined for GI as a function of all continuous life-history and physiological traits , including adult cortical neuron counts . For three eulipotyphla species ( Sorex fumeus , Blarina brevicauda , Scalopus aquaticus ) , data were available for neuron counts but not GI , and therefore we extrapolated the GI of those species on the basis of gross morphology . Finally , to test whether the bimodal distribution of GI may be influenced by the topology of the mammalian phylogenetic tree , we used an expectation-maximization algorithm . Each simulated trait was given the same variance as GI ( Figure S5 ) and the result was averaged over 104 simulated datasets . None of the simulations produced the same bimodal distribution of species observed for GI data . We used a comprehensive phylogenetic approach to map 37 life-history and physiological character traits collected from the literature ( Tables S1 and S2 ) onto hypotheses of phylogenetic relationships in Mammalia , in order to examine how those traits correlate , over evolutionary time , with degree of gyrencephaly . Continuous character traits were discretized using the consensus of natural distribution breaks calculated with a Jenks-Caspall algorithm [56] , model-based clustering according to the Schwarz criterion [57] , and hierarchical clustering [58] . Character histories were then corrected for body mass with a phylogenetic size correction [59] , [60] and summarized across the phylogeny using posterior probabilities . Associations between individual states of each character trait along those phylogenetic histories were calculated in SIMMAP ( v1 . 5 ) using empirical priors based on the frequency of character states for each trait [61]; the association between any two states was a measure of the frequency of occurrence ( i . e . , the amount of branch length across the tree ) of those states on the phylogeny . While correcting for body mass is intended to normalize the data , it cannot completely remove interdependencies between character traits . Although we cannot a priori assume that any of the traits interact , exploring interactions between them deserves further investigation . The sums , rates , and types of changes for GI and body weight were plotted as mutational maps to assess directional biases in their evolution [62]–[64] . These were used to determine the evolutionary historical patterns of GI and , as a control , body weight . By estimating the occurrence ( number of times an increase/decrease happens ) and timing ( where in the phylogeny the change occurs ) of different values for each trait , we were able to calculate how often each trait has increased and decreased in mammalian evolution . We were therefore able to evaluate the ratio of increases over decreases for each trait ( Figure S4 ) . We estimated neuroepithelial founder pool populations for mouse and human . For the mouse , we used coronal sections of an E11 . 5 mouse embryo obtained from the Allen Brain Atlas [65] . We obtained 19 sections equidistantly spaced along the anterior-posterior axis of the brain . The length of the ventricular surface of the dorsal telencephalon was manually traced in Fiji [66] on each section starting from the point above the nascent hippocampus and ending in the point above the lateral ganglionic eminence . The horizontal length of the embryonic brain at E11 . 5 was measured with images from [67] . Using the coronal and horizontal measurements , we constructed a polygon representing the ventricular surface of the dorsal telencephalon and calculated the area of this surface in Fiji . We measured the surface area of the end-feet of neuroepithelial cells using EM images of the coronally cut apical surface of an E11 . 5 embryonic mouse brain ( Table S6 ) . The diameter of a single end-foot was calculated by measuring the distance between the adherens junctions . We corroborated these end-feet calculations with published immunofluorescence stainings of the apical complex ( ZO1 and N-cadherin ) from an en face perspective [68] , [69] . The average surface area of a single end-foot was calculated by approximating the end-foot as a hexagon; and the number of founder cells was estimated by dividing the surface of the dorsal telencephalon by the surface of an individual end-foot of the neuroepithelial cell , such thatOur final mouse values were comparable to those previously published [70] . For the human , we followed the same procedure , using ten coronal sections and one horizontal section of a gestation week ( GW ) 9 brain [71] . End-feet were calculated using EM images of the apical surface of a human brain at GW13 . The measurements are available in Table S6 . Because the number of founder cells per surface area was nearly equivalent in mouse and human ( 4×105/mm2 ) , we used this ratio , along with data on ventricular volume collected from the literature ( Tables S1 and S2; Database S1 ) , to estimate neuroepithelial founder cell populations for a further 15 species ( Table 1 ) . For species where no data on ventricular volume were available , values were estimated on the basis of a regression analysis against brain weight ( Figure S6 ) . Ventricular volume was then converted to surface area for each species by approximating the ventricle as a cylinder with a 4 . 5-to-1 height-to-diameter proportion ( this ratio was estimated on the basis of observations in mouse ) . Ventricular volume-derived ventricular surface area estimates were corroborated with the surface areas calculated from the literature for mouse and human . Founder cell estimates were then computed on the basis of the densities derived above for mouse and human . Using this method , but alternately ignoring our mouse and human calculations to define the parameters , we were able to predict mouse and human values within 10% of our calculations , respectively . Workers have demonstrated the occurrence of three primary lineages of neuron generation in mouse corticogenesis ( Figure 5A , lineages 1–3 ) [1] , [5] , [14] , [72] and a further four lineages in primate corticogenesis ( Figure 5A , lineages 4–7 ) [9] , [10] . While there is evidence for at least one additional lineage in mouse [6] , and further lineages may be speculated , we limited our model to the seven that are considered to contribute most significantly to neuron output [2] , [10] , [73] , [74] . The extent of neuron generation in each of these seven lineages was summarized in series and solved numerically ( Figure 5B ) . Neurogenic period was either taken from the literature ( Database S1 ) or estimated on the basis of a regression analysis of neurogenic period as a function of gestation period ( Figure S6 ) . Neurogenic period in human was estimated using empirical observations from the literature [75]–[77] . The averaged cell-cycle length for apical and BPs from the mouse ( 18 . 5 hours ) [78]–[80] was used for all non-primates; averaged cell-cycle length for cortical areas 17 and 18 from the macaque ( 45 hours ) was used for catarrhines [10] , [81]; and an intermediary cell-cycle length ( 30 hours in marmoset , determined by EdU labeling; Ayako Murayama and Hideyuki Okano , personal communication ) , was used for platyrrhines . Using an average cell-cycle length value for all progenitor-types was found to be equally valid for predicting neuron number as using different cell-cycle lengths for each progenitor-type ( Figure S12 ) . Therefore , despite its potential shortcomings , using average cell-cycle length is a valid approach and , given the scarcity of species data on the cell-cycle length of various progenitor-types at different stages of neurogenesis , the best approach available to construct neurogenic models across many species . Generous confidence intervals ( 75% ) for cell-cycle length are used in our models ( Figure 5C ) , in order to show the minimal explanatory power cell-cycle length provides for interspecific differences in cortical neuron number . Diminishing numbers of neuroepithelial cells have been observed to continue to proliferate at the ventricle until E18 . 5 in the mouse [7] . Therefore , final neuroepithelial founder pool estimates were calculated from the aforementioned by evenly decreasing the value of n in the Sherley equation [82] from 1 at E9 . 5 to 0 at E18 . 5 in the mouse and at comparable neurogenic stages in other species . Neuron numbers were calculated for each species from combinations of lineages . The proportional contribution of each lineage for each species was parameterized according to existing data on progenitor cell-type abundances in mouse [14] , marmoset [22] , rabbit ( IK and WBH , in preparation ) , macaque [10] , and human [9] , [11] . Where no such data were available , proportional contributions were permutated for all lineages until a best-fit estimate , based on cortical neuron numbers taken from the literature [33] , [83]–[85] , was achieved ( Tables 1 and 2 ) . Each lineage was assumed to occur from the first to final day of neurogenesis , although this is only approximately accurate . Finally , because of published estimates of postnatal apoptosis in the mammalian cortex [86]–[88] , we assumed neuron counts to be 1 . 5-fold higher at the termination of neurogenesis than in the adult brain; therefore , neuron number at the termination of neurogenesis was estimated in each species by multiplying neuron numbers collected from the literature by 1 . 5 . This multiplication is not represented in Table 1 . Trade-offs in adapting a human neurogenic program with either an expanding neuroepithelial founder pool or lengthening neurogenic period were tested for the mouse ( Mus musculus ) and marmoset ( Callithrix jacchus ) , two lissencephalic species whose cell-type proportions during neurogenesis have been documented [7] , [14] , [22] . To estimate the relative reproductive value and stable-stage proportions of each of the lineages in the mouse and human neurogenic programs , we constructed a stage-structured Lefkovitch matrix , using sums of the lineage series ( after 100 cycles ) as fecundity values and complete permutations of the proportional contributions of each lineage as mortality values . The altered growth-rates of each lineage were calculated by excluding lineages one at a time and assuming 100% survival in the remaining lineages ( Table S5 ) . We introduced three ODEs to explore the average dynamics of asymmetric versus symmetric progenitors , such that: if a ( t ) , b ( t ) , and c ( t ) are the numbers of asymmetrically dividing cells , differentiated cells , and proliferative cells , respectively , then , where r is equal to growth-rate . If a ( t ) = a0 , thenandUsing these ODEs , we calculated the effect on neuron output of increasing the likelihood of symmetrically dividing daughter progenitors in the lineage ( Figure S9 ) . The interdependent growth-rates in the model reflect a purely mechanistic interpretation of determining neuronal output from a finite pool of asymmetrically dividing cells . The ODEs , therefore , may not reflect differential regulation of neuronal output via direct versus indirect neurogenesis . The daughter proliferative cells are designed to carry out one round of proliferation followed by a final round of self-consumption ( Figure S9 ) .
|
What are the key differences in the development and evolution of the cerebral cortex that underlie the differences in its size and degree of folding across mammals ? Here , we present phylogenetic evidence that the Jurassic era mammalian ancestor may have been a relatively large-brained species with a folded neocortex . We then show that variation in the degree of cortical folding ( gyrencephaly index [GI] ) does not evolve linearly across species , as previously assumed , but that mammals fall into two principal groups associated with distinct ecological niches: low-GI mammals ( such as mice and tarsiers ) and high-GI mammals ( such as dolphins and humans ) , which are found to generate on average 14-fold more brain weight per day of gestation . This greater daily brain weight production in mammals with a highly folded neocortex requires a specific class of progenitor cell-type to adopt a special mode of cell division , which is absent in mammals with slightly folded or unfolded neocortices . Differences among mammals within the same GI group ( high or low ) are not due to different programming , but rather the result of differences in the length of the neurogenic period . So , the impressively large and folded human neocortex , which is three times the size of the chimpanzee neocortex , can be explained by a modest evolutionary extension of the neurogenic period with respect to its closest primate ancestors .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"developmental",
"neuroscience",
"cellular",
"neuroscience",
"neurogenesis",
"ecology",
"simulation",
"and",
"modeling",
"mathematical",
"and",
"statistical",
"techniques",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"neuroscience",
"research",
"and",
"analysis",
"methods"
] |
2014
|
An Adaptive Threshold in Mammalian Neocortical Evolution
|
The host limits adenovirus infections by mobilizing immune systems directed against infected cells that also represent major barriers to clinical use of adenoviral vectors . Adenovirus early transcription units encode a number of products capable of thwarting antiviral immune responses by co-opting host cell pathways . Although the EGF receptor ( EGFR ) was a known target for the early region 3 ( E3 ) RIDα protein encoded by nonpathogenic group C adenoviruses , the functional role of this host-pathogen interaction was unknown . Here we report that incoming viral particles triggered a robust , stress-induced pathway of EGFR trafficking and signaling prior to viral gene expression in epithelial target cells . EGFRs activated by stress of adenoviral infection regulated signaling by the NFκB family of transcription factors , which is known to have a critical role in the host innate immune response to infectious adenoviruses and adenovirus vectors . We found that the NFκB p65 subunit was phosphorylated at Thr254 , shown previously by other investigators to be associated with enhanced nuclear stability and gene transcription , by a mechanism that was attributable to ligand-independent EGFR tyrosine kinase activity . Our results indicated that the adenoviral RIDα protein terminated this pathway by co-opting the host adaptor protein Alix required for sorting stress-exposed EGFRs in multivesicular endosomes , and promoting endosome-lysosome fusion independent of the small GTPase Rab7 , in infected cells . Furthermore RIDα expression was sufficient to down-regulate the same EGFR/NFκB signaling axis in a previously characterized stress-activated EGFR trafficking pathway induced by treatment with the pro-inflammatory cytokine TNF-α . We also found that cell stress activated additional EGFR signaling cascades through the Gab1 adaptor protein that may have unappreciated roles in the adenoviral life cycle . Similar to other E3 proteins , RIDα is not conserved in adenovirus serotypes associated with potentially severe disease , suggesting stress-activated EGFR signaling may contribute to adenovirus virulence .
Human adenoviruses provide an excellent example of how viruses adapt host cell machinery to invade cells , gain access to the nucleus to replicate , assemble new viral particles , and spread in the host [1 , 2 , 3] . The host limits adenovirus infections by mobilizing innate immune systems that activate inflammatory or cytotoxic responses directed against infected cells [3 , 4 , 5 , 6] . These host defense mechanisms also represent a major barrier to the use of adenovirus vectors with many important clinical applications , ranging from cancer gene therapy to vaccine development [3 , 4 , 5 , 6] . In addition to specialized immune cells that secrete pro-inflammatory cytokines at sites of infection , immune and non-immune target cells both rely on cell autonomous innate immunity to defend against the immediate threat of infection [7] . Adenovirus circumvents various innate defense mechanisms by virtue of viral proteins encoded by early transcription units that strike a balance between the elimination of virus and immune-mediated tissue injury [8] . The study of cellular pathways used by viruses has led to many significant advances in eukaryotic cell and molecular biology [9] . Adenovirus early region 3 ( E3 ) gene products in particular have been powerful tools for discovering new mechanisms in the fields of intracellular protein and lipid trafficking [10 , 11 , 12] . The adenovirus gene regulatory program involves two distinct phases during lytic infections . E3 transcripts are maximally expressed through transactivation by the early region 1A ( E1A ) gene product during the early phase , and subsequently repressed after the onset of DNA replication [13 , 14] . The E3 promoter also has NFκB binding sites that are highly sensitive to activation signals in human T cells in the absence of E1A , suggesting these proteins have key roles in persistently infected T lymphocytes [15] . The E3 region encodes several proteins that are specifically involved in immune evasion by modulating the function of cell surface receptors , intracellular signaling events , and secretion of pro-inflammatory molecules [11 , 16 , 17 , 18] . Many E3 proteins target host cell pathways with important roles in both immune and non-immune cells [11 , 16 , 17 , 18] . However , other host targets such as the EGF receptor ( EGFR ) are predominantly expressed in non-immune epithelial cells , suggesting E3 proteins also have cell-specific functions that have not been fully explored [19] . Although most adenovirus infections are self-resolving , some serotypes can lead to a serious and frequently fatal condition called acute respiratory distress syndrome [3 , 4 , 20] . Over the past few years adenoviruses have increasingly been recognized as significant pathogens associated with high morbidity and mortality in immune-compromised individuals , especially in the pediatric hematopoietic transplant population [21 , 22 , 23 , 24 , 25 , 26] . However , periodic adenovirus outbreaks can also pose significant health risks in people with no known predisposing conditions , such as US military personnel and civilians in closed community settings [27 , 28] . Understanding the cellular mechanisms targeted by adenoviruses is essential for developing new therapeutic approaches against viral diseases . In addition , common human diseases such as cancer and diabetes frequently manipulate the same cellular pathways as infectious agents [29 , 30] . In contrast to E1A and early region 1B ( E1B ) genes that are conserved among different adenovirus serotypes , E3 genes vary markedly and may even be absent , supporting their potential roles in adenovirus pathogenesis [26] . The inducible expression of pro-inflammatory cytokines under regulation of the nuclear factor kappa-light-chain-enhancer of activated B cells ( NFκB ) family of transcription factors is a key component of the host innate immune response to infectious adenoviruses and adenovirus vectors [3 , 4 , 31] . As with other viruses such as HSV-1 , NFκB activation appears to occur in waves that presumably induce different patterns of gene expression over the course of an acute adenovirus infection [32] . The first wave is regulated by a rapid , transient mechanism involving PI3K ( phosphoinositide 3-kinase ) /Akt signaling , which is activated by the interaction between adenoviral capsids and host cell alpha ( V ) integrin receptors in pure cultures of alveolar epithelial cells [33 , 34 , 35 , 36 , 37] . A second wave has been linked to early gene expression . For instance , multiple lines of evidence indicate that E1A sensitizes infected respiratory epithelial cells to bacterially-derived LPS ( lipopolysaccharide ) , which is present as a contaminant on airborne particles and activates NFκB following its binding to TLR4 ( Toll-like receptor 4 ) [38] . A second example involves the E3-19K viral protein , which is known to trigger protein overload in the endoplasmic reticulum ( ER ) leading to calcium release and subsequent production of reactive oxygen intermediates mediating NFκB activation [39] . Mechanisms for switching between different modes of NFκB-dependent gene transcription are not currently known . It has also not yet been established how factors such as cellular stress shape the NFκB response to adenoviral infection [40 , 41] . Recent evidence indicates that a variety of cellular stresses or stress inducers stimulate a robust , non-canonical pathway of ligand-independent EGFR trafficking , frequently downstream of p38-MAPK activity [42 , 43] . Although these stress-induced EGFR responses are thought to provide cancer cells with a survival advantage and resistance to therapeutics , their potential roles during viral infections are largely unknown [42 , 43] . In contrast to ligand-stimulated counterparts that are targeted for degradation , stress-exposed EGFRs accumulate on both limiting membranes and intraluminal vesicles ( ILVs ) in a relatively stable population of multivesicular body ( MVB ) endosomes where they are subsequently activated [42 , 43] . ILVs appear to undergo dynamic cycles of fission and fusion at the limiting membrane , enabling resumption of EGFR stress signaling from MVB limiting membranes [44] . It is also thought that ILV back-fusion facilitates EGFR recycling back to plasma membrane when p38-MAPK activity declines [45] . A number of EGFR signaling pathways have been linked to NFκB activation downstream of ligand stimulation or constitutive EGFR activation in cancer cells , primarily by promoting degradation of inhibitor of kappa B ( IkB ) proteins that sequester inactive NFκB proteins in the cytoplasm [40 , 46] . However , there is little information regarding potential links between stress-induced EGFR signaling and NFκB . We have reported previously that cell surface levels of EGFR and related tyrosine kinase family members were significantly reduced following acute infection with group C adenoviruses [19 , 47 , 48] . Using an extensive series of adenovirus deletion mutants , we mapped the viral gene responsible for this effect to an E3 transcript encoding a small membrane protein called RIDα-C2 [48 , 49] . We have also shown that RIDα-C2 was a resident membrane protein in a novel population of endosomes , where it transiently interacted with EGFRs and re-routed them for degradation [19 , 50 , 51 , 52] . In contrast to the ligand-induced pathway , however , adenovirus-induced EGFR trafficking did not require intrinsic tyrosine kinase activity or receptor ubiquitination [19 , 53] . However , the role of EGFR in adenovirus pathogenesis has remained elusive , due in part to an incomplete understanding of the underlying mechanism of RIDα-induced EGFR trafficking . Here we report two novel findings . First , adenovirus infection induced an EGFR stress response that generated a phosphorylated NFκB binding site for the peptidyl-prolyl isomerase Pin1 , which is known to enhance NFκB activity by countering negative feedback control through ubiquitin ( Ub ) -mediated NFκB degradation [54] . Second , the adenoviral RIDα protein terminated this pathway by co-opting host machinery regulating sorting of stress-exposed EGFRs in MVB endosomes and promoting lysosome fusion . These studies have yielded new insights to the molecular basis of stress-induced EGFR trafficking and signaling , as well as unappreciated EGFR functions that may be exploited by other viruses .
A majority of our previous studies analyzing the effect of adenovirus infection on EGFR trafficking were carried out in cancer cell lines and heterologous cell models with pathological levels of EGFR ( > 106 receptors/cell ) [19 , 51 , 53 , 55] . In order to establish the role of host proteins in adenovirus-regulated EGFR trafficking , it was important to utilize cells with receptor expression closer to physiological levels to avoid saturation of the trafficking machinery [56 , 57 , 58] . We tested whether A549 lung epithelial cells were an appropriate tissue culture model for two main reasons . First , although this is a cancer cell line , A549 cells express wild-type ( WT ) EGFRs within the range of physiological receptor expression ( ~105 receptors/cell ) [55 , 59] . Second , they are a well-established airway epithelial tissue culture model for analyzing host responses to acute adenovirus infections and adenoviral-based therapeutics [35 , 60 , 61 , 62] . To determine their suitability for these studies , A549 cells were pulse-labeled with radioactive amino acids shortly after infection with HAdV-C2 , and RIDα and EGFR proteins were recovered by immunoprecipitation for analysis by SDS-PAGE . We found that the viral protein reached steady-state expression by 6 h post-infection ( p . i . ) ( Fig 1A ) . HAdV-C2 infection was also associated with a significant reduction in EGFR metabolic half-life , which typically ranges from 18 to 24 h under basal conditions [63 , 64] ( Fig 1A and 1B ) . Total EGFR protein levels were compared by immunoblotting whole cell lysates from cells infected with HAdV-C2 , versus a HAdV-C2 mutant with an amino-terminal 107 base pair deletion in the RIDα gene ( labeled “RIDα-null” ) [49 , 65] . Cells infected with either virus expressed equivalent levels of E1A , the first viral gene product expressed post-infection ( Fig 1C ) . However , only cells infected with HAdV-C2 produced E3-encoded RIDα proteins ( Fig 1C ) . Consistent with reports in the literature , HAdV-C2 infection was associated with a significant reduction in total tumor necrosis factor receptor ( TNFR1 ) protein , but did not significantly alter total protein levels of the transferrin receptor ( TfR ) ( Fig 1C ) . In contrast to TNFR1 , infection with the RIDα-null virus was sufficient to block EGFR down-regulation independently of other virally-encoded proteins ( Fig 1C ) . Confocal imaging revealed that EGFR accumulated in intracellular vesicles following infection with HAdV-C2 , in contrast to mock-treated cells where EGFRs were predominantly associated with plasma membrane ( Fig 1D ) . Internalized EGFRs were extensively co-localized with markers for early ( EEA1 and Hrs ) endosomes at 8 to 10 h post-infection , suggesting EGFRs trafficked to these compartments prior to degradation ( Fig 1E ) . EGFR distribution on MVB limiting membranes and ILVs was examined by transmission electron microscopy in cells infected with HAdV-C2 or the RIDα-null mutant virus , that were stained with a colloidal gold-conjugated EGFR antibody ( Fig 1F ) [66] . Gold particles associated with MVB compartments were quantified in a total of 50 cells in two independent experiments . We found that MVB limiting membranes had similar EGFR content in cells infected with either virus , but that there was a statistically significant increase in the number of gold particles localized on MVB ILVs in cells infected with HAdV-C2 compared to the RIDα-null mutant virus ( Fig 1F ) . Collectively these data established that RIDα promoted endolysosomal EGFR sorting at physiological levels of receptor expression in HAdV-C2 infected epithelial cells . For the majority of membrane proteins that have been studied in detail including ligand-activated EGFRs , ILV sorting is regulated by the canonical ESCRT machinery targeting Ub-cargo ( Fig 2A ) [67] . Briefly , ESCRT-0 comprised of Hrs and Stam1 subunits sequesters ligand-stimulated receptors in peripheral early endosomes , and ESCRT-I and ESCRT-II act sequentially to sort Ub-EGFRs into inward invaginations of limiting membranes [68] . ESCRT-III then facilitates cargo deubiquitination and membrane scission finalizing ILV formation [69] . The molecular basis of the adenovirus-induced sorting pathway was determined by systematic depletion of key subunits italicized in Fig 2A , which are known to regulate the functional integrity of various ESCRT complexes , using targeted small interfering RNAs ( siRNAs ) ( Fig 2B and 2C ) [70] . Depletion of individual ESCRT subunits led to a significant reduction in ligand-induced EGFR down-regulation ( Fig 2D and 2E ) , consistent with reports in the literature [70] . In contrast , the adenovirus-induced pathway was inhibited by reduced ESCRT-0 function , but ESCRT-I and ESCRT-II were dispensable ( Fig 2F and 2G ) . The lack of phenotype was not due to differences in infectivity , since E1A proteins were similarly expressed in cells with reduced expression of individual ESCRT subunits ( Fig 2F ) . These findings suggested that adenovirus-regulated EGFR trafficking had overlapping but distinct requirements for ESCRT machinery compared to the ligand-stimulated pathway . They are also consistent with previous results that receptor ubiquitination was not required for adenovirus-mediated EGFR trafficking [53] . The multi-adaptor protein Alix has recently emerged as a key component in alternative ILV sorting pathways because of its ability to simultaneously bind non-Ub cargo , and mediate ILV sorting via a direct protein-protein interaction with the ESCRT-III subunit CHMP4B [71] . The role of Alix in HAdV-C2 infected cells was evaluated by siRNA gene silencing , which reduced Alix expression by ~ 80% relative to a non-coding siRNA ( Fig 3A ) . As reported by several other groups , Alix depletion did not have a significant impact on EGFR down-regulation following ligand stimulation ( Fig 3B and 3C ) [72 , 73 , 74 , 75 , 76] . In sharp contrast , Alix gene silencing effectively blocked EGFR degradation in the HAdV-C2 induced pathway , without impacting viral infectivity assessed by E1A expression ( Fig 3D and 3E ) . We also showed that Alix was detected in RIDα immune complexes isolated from cells with physiological levels of endogenous Alix and stable expression of a FLAG-tagged RIDα protein encoded by HAdV-C2 ( RIDα-C2 ) , suggesting these two proteins interacted under physiological conditions in vitro ( Fig 3F ) . This interaction was also evaluated by incubating whole cell lysates with GST fusion proteins containing the 30-amino acid cytosolic tail of RIDα-C2 and truncated RIDα-C2 peptides with premature stop codons ( Fig 3G and 3H ) [52 , 77] . Immunoblot analysis revealed that RIDα-C2 and Alix formed a molecular complex involving RIDα-C2 residues 77-PQYR-80 , which were located adjacent to a previously described binding site for the Rab7 effector RILP ( Rab7 interacting lysosomal protein ) ( Fig 3G and 3H ) [77] . Collectively our data supported a working model that HAdV-C2 diverted EGFRs to a non-canonical degradative pathway , via a direct interaction between a proline-containing motif in the RIDα cytosolic tail and Alix . It was also notable that HAdV-C2 co-opted the ESCRT-0 subunit Hrs ( Fig 2F and 2G ) and Alix ( Fig 3 ) , since both of these ESCRT components are required for stress-induced EGFR trafficking [44] . The hypothesis that RIDα modulated the stress-induced EGFR trafficking pathway was tested in cells treated with the pro-inflammatory cytokine TNF-α . The initial set of experiments confirmed that TNF-α elicited a typical EGFR stress response in A549 cells . We first showed that TNF-α induced EGFR phosphorylation at Ser1046/1047 , which is known to mediate stress-induced EGFR internalization , in A549 cells ( Fig 4A ) [42 , 78] . In contrast , EGF stimulated EGFR phosphorylation at Tyr1045 that is the docking site for the E3 Ub ligase c-Cbl , which was also tyrosine phosphorylated in EGF-stimulated cells ( Fig 4A ) . In addition , EGFR ubiquitination required in the canonical ESCRT pathway was induced by EGF but not TNF-α ( Fig 4B ) . Finally , we demonstrated that internalized EGFRs accumulated in EEA1-positive early endosomes following 60-min stimulation with TNF-α ( Fig 4C ) . EGFR stress responses were then compared in parent A549 cells versus cells with stable expression of FLAG-tagged RIDα-C2 . We first showed that the viral protein did not have a significant impact on steady-state protein expression of the TNF-α receptor TNFR1 ( Fig 4D ) . TNF-α also induced rapid p38-MAPK activation and EGFR phosphorylation at Ser1046/47 in parent A549 cells and cells with stable RIDα expression ( Fig 4E ) . In addition , TNF-α stimulated EGFR activation in both cell models , as determined by immunoblotting with a phospho-specific EGFR antibody to the Tyr1068 autophosphorylation site ( Fig 4F ) . However , autophosphorylated and total EGFR protein levels were markedly reduced within 2 to 3 h of TNF-α addition to RIDα-C2 expressing cells , compared to parent cells ( Fig 4F ) . Furthermore , reduced EGFR expression observed in A549 + RIDα-C2 cells was blocked when cells were pretreated with the lysosomotropic agent chloroquine to inhibit protein degradation in lysosomes ( Fig 4G ) [79] . Altogether these results indicated that RIDα expression was sufficient to divert stress-internalized EGFRs to lysosomes . Site-specific phosphorylation events play a critical role in the regulation of NFκB-p65 activity downstream of multiple stimuli , including TNF-α [80] . While many of the TNF-α-induced upstream signals have been identified , the pathway responsible for phosphorylation at the Thr254-Pro motif , which is a known substrate for the peptidyl-prolyl isomerase Pin1 , was unknown [54] . We therefore used a genetic approach to investigate whether stress-activated EGFR contributed to phosphorylation at Thr254-Pro , by comparing responses in EGFR-null mouse fibroblasts versus cells that were reconstituted with elevated levels of wild-type ( WT ) human EGFR . We first showed that EGFR was autophosphorylated at Tyr1068 in response to TNF-α stimulation in the reconstituted cells ( Fig 5A ) . The EGFR activation profile appeared to be biphasic , which would be consistent with re-activation of EGFRs that have recycled from ILVs back to MVB limiting membranes [42 , 43] . TNF-α stimulated phosphorylation at Ser468 located in the NFκB-p65 transactivation domain with similar kinetics and duration in both cell lines ( Fig 5A ) . However , robust Thr254 phosphorylation was only observed in the EGFR-reconstituted cells ( Fig 5A ) . It has already been established that Pin1-mediated prolyl isomerization leads to NFκB-p65 nuclear accumulation , increased NFκB-p65 protein stability , and enhanced transcription [54 , 81] . We therefore tested the hypothesis that stable RIDα expression would have a negative impact on TNF-α-induced NFκB-p65 signaling . In cell fractionation studies , we showed that Ser468-phosphorylated and Thr254-phosphorylated NFκB-p65 species were mainly localized in the nucleus , consistent with nuclear translocation , in both cell models ( Fig 5B ) . However , nuclear levels of total and phosphorylated NFkB-p65 proteins were markedly reduced in cells expressing the RIDα-C2 protein ( Fig 5B ) . We also examined production of the NFκB target interleukin-8 ( IL-8 ) by ELISA of tissue culture supernatants ( Fig 5C ) . The induction of IL-8 expression 24 h post-stimulation ( ~ 0 . 6 ng/ml ) was on par with other reports for TNF-α treatment of A549 cells ( e . g . [82] ) ( Fig 5C ) . Constitutive RIDα-C2 expression led to a significant reduction in secreted IL-8 protein ( Fig 5C ) , consistent with attenuated NFκB-p65 gene transcription . Finally , we showed that stress-activated EGFR signaling through the adaptor protein Gab1 was also attenuated in the RIDα-C2 expressing cells compared to parent A549 cells ( Fig 5D ) . These results suggested that the EGFR/NFκB pathway may be part of a broader stress-induced signaling network . Although NFκB-p65 was known to be activated by viral cell entry in alveolar epithelial cells , the upstream signaling mechanisms regulating this important pro-inflammatory pathway are incompletely understood [33 , 35 , 36 , 37] . Based on results in TNF-α-stimulated cells , we tested whether NFκB-p65 phosphorylation was regulated by EGFR signaling as a stress response to HAdV-C2 infection . Confirming reports in the literature , p38-MAPK was rapidly activated in cells infected with HAdV-C2 at a multiplicity of infection ( MOI ) of ~100 particles/cell ( Fig 6A ) [83] . HAdV-C2 infection also led to rapid EGFR phosphorylation at residues Ser1046/1047 , which are substrates for the p38-MAPK effector MK2 ( mitogen-activated protein kinase-activated protein kinase 2 ) that is known to be activated by viral cell entry ( Fig 6A ) [83 , 84 , 85 , 86] . We also found that EGFR was autophosphorylated at Tyr1068 several hours before the adenovirus E1A gene product was expressed ( Fig 6B ) . Similar to results observed in TNF-α-stimulated cells ( see Fig 5A ) , the EGFR activation profile was biphasic ( Fig 6B and 6C ) . In addition , HAdV-C2 infection triggered NFκB-p65 Thr254 phosphorylation prior to E1A gene product detection by immunoblotting ( Fig 6B ) [54 , 87] . To rigorously rule out a role for the RIDα protein in these early responses , we showed that the RIDα-null mutant virus led to the same site-specific EGFR and NFκB-p65 phosphorylation events within the first 90-min of infection ( Fig 6D ) . Since conventional tyrosine kinase inhibitors may not be effective towards transactivated EGFRs [43] , we used a genetic approach to verify a role for stress-induced EGFR activity in NFκB-p65 signaling induced by adenovirus infection . These studies were carried out in EGFR-null mouse fibroblasts that were reconstituted with WT-EGFR , or a kinase-dead EGFR mutant that we had shown previously was down-regulated in adenovirus-infected cells similar to the WT receptor [19] . Although both receptor proteins were phosphorylated at Ser1046/1047 shortly after viral infection , EGFR underwent stress-induced autophosphorylation in cells expressing WT but not kinase dead-EGFR ( Fig 6E ) . In addition , the phospho-Thr254 NFκB-p65 modification was only observed in infected cells expressing WT-EGFR , supporting the hypothesis that EGFR signaling was the upstream pathway regulating Thr254 phosphorylation ( Fig 6E ) . Our previous studies have shown that RIDα-C2 has several small interaction modules composed of 2–6 residues contributing to various regulatory functions located within its C-terminal region ( Fig 7A ) . In addition to binding motifs for Alix and RILP that have already been described , RIDα-C2 interacts with the clathrin adaptor AP1 regulating its trafficking from the trans-Golgi network to endosomes; and the oxysterol binding protein ORP1L responsible for coupling the viral protein to homeostatic regulatory sterol pools in the ER [52 , 77 , 88] . RIDα-C2 function has also been shown to be regulated by reversible palmitoylation at Cys67 [89] . Interestingly , pathogenic adenovirus serotypes HAdV-B7 and HAdV-E4 differ in two regards: They lack the palmitoylation site , and the Alix binding site has negative/positive to uncharged amino acid substitutions ( Fig 7A ) . The goal of these experiments was to determine whether these divergent sequences were associated with differences in the stress-activated EGFR/NFκB-p65 signaling pathway . Similar to HAdV-C2 ( see Fig 6A and 6B ) , HAdV-B7 and HAdV-E4 both triggered EGFR autophosphorylation at Tyr1068 , and NFκB-p65 phosphorylation at Thr254 , within the first hour of infection ( Fig 7B ) . We next showed that HAdV-B7 encoded an immunologically cross-reactive RIDα protein , but failed to significantly reduce EGFR metabolic stability in contrast to HAdV-C2 ( Fig 7C ) . Similarly , HAdV-B7 and HAdV-E4 did not have a discernible effect on total EGFR protein compared to HAdV-C2 , following an 18 h infection ( Fig 7D ) . Cells infected with HAdV-B7 and HAdV-E4 also exhibited increased nuclear accumulation of NFκB-p65 protein relative to HAdV-C2-infected cells ( Fig 7E ) . Similar to results using TNF-α as a stress stimulus , NFκB-p65 nuclear accumulation was attenuated when A549 cells with constitutive RIDα-C2 expression were infected with HAdV-B7 ( Fig 7E ) . We also found that HAdV-B7 triggered a significant increase in IL-8 expression that was on par with other reports in the literature ( ~ 1 . 2 ng/ml 24 h post-infection ) compared to HAdV-C2 ( Fig 7F ) [90] . Furthermore , IL-8 production was significantly reduced in cells with constitutive RIDα-C2 expression infected with either HAdV-C2 ( **P < 0 . 0001 ) or HAdV-B7 ( *P = 0 . 0206 ) ( Fig 7F ) . Our results supported a hypothesis that stress-induced EGFR signaling , and its serotype-specific antagonism in HAdV-C2 , are important factors in shaping the epithelial cell response to adenovirus infections . Stress-internalized EGFRs generally recycle back to the plasma membrane when the cellular stress has resolved , implying that the RIDα protein changes the balance between EGFR recycling and degradation during stress-induced receptor trafficking [42 , 43] . EGFR down-regulation induced by HAdV-C2 was not impaired by Rab7 gene silencing , indicating EGFRs were not diverted to the canonical Rab7 degradative pathway in infected cells ( Fig 8A ) . This finding was consistent with our previous results showing that adenovirus-mediated EGFR trafficking was regulated by an interaction between RIDα-C2 and RILP , which couples Rab7 to the HOPS ( homotypic fusion and vacuole protein sorting ) complex responsible for late endosome-lysosome fusion ( Fig 8B ) [77] . To determine whether the viral protein induced lysosomal EGFR sorting in the absence of other viral proteins , we tested whether constitutive RIDα expression rescued ligand-induced EGFR down-regulation in Rab7-depleted cells . Rab7 gene silencing effectively blocked ligand-induced down-regulation in parent A549 cells , but not in cells with stable RIDα-C2 expression ( Fig 8C and 8D ) . In contrast , the RIDα-B7 protein encoded by HAdV-B7 did not reconstitute ligand-induced EGFR trafficking in Rab7-depleted cells ( Fig 8C and 8D ) . Results in A549 cells expressing different RIDα proteins therefore supported the hypothesis that serotype-specific amino acid sequences ( see Fig 7A ) are key determinants of functional activity regulating EGFR trafficking . We also observed a time-dependent reduction in EGFR protein that could be extracted with non-ionic detergent , following Rab7 depletion in parent A549 cells and cells reconstituted with RIDα-B7 ( arrows in Fig 8D ) . Although the reasons for this are not clear , we speculate that internalized EGFRs may be mis-trafficked through aberrant compartments with accumulated lipid raft components , which are not fully solubilized with non-ionic detergents [91 , 92] . Rab7 also regulates autophagosome maturation through a different effector molecule linking it to the HOPS tethering complex ( Fig 8B ) . We therefore determined whether RIDα reconstituted starvation-induced autophagy in Rab7-depleted cells . Autophagic flux was assessed by examining turnover of microtubule-associated protein light chain 3 ( LC3 ) , which is converted from its cytosolic form ( LC3-I ) to a lipid-modified species ( LC3-II ) and subsequently degraded during autophagic progression [93 , 94 , 95 , 96 , 97] . Rab7 depletion was associated with a build-up of LC3-II in amino acid starved cells consistent with a block in autophagosome-lysosome fusion ( Fig 8E ) . Rab7-depleted cells also accumulated the adaptor protein p62 that acts as a cargo receptor for targeting Ub-substrates to autophagosomes [98] ( Fig 8E ) . In contrast to ligand-induced EGFR turnover , however , heterologous RIDα expression did not reconstitute autophagic flux of LC3 or p62 in Rab7-depleted cells ( Fig 8D ) . In fact , stable RIDα expression led to a modest reduction in autophagic flux even under nutrient-rich conditions ( Fig 8E ) . These results suggested RIDα-C2 altered the trafficking fate of stress-internalized EGFRs by mimicking Rab7-regulated lysosome fusion with late endosomes but not autophagosomes .
Our studies have revealed three major new findings . First , incoming viral particles induced a non-canonical pathway of stress-activated EGFR trafficking and signaling prior to nuclear translocation and transcription of viral DNA in epithelial target cells . Second , stress-induced EGFR signaling was required for a site-specific phosphorylation event with a pivotal role in NFκB-p65 function . Third , the adenoviral RIDα protein attenuated the EGFR/NFκB signaling axis in the context of an acute adenovirus infection , and as an independently expressed transgene in cells stimulated with the pro-inflammatory cytokine TNF-α . The following working model interprets data presented here in the context of the current literature ( see Fig 9 for a schematic diagram ) . The first set of events in the working model depicts the known contribution of consecutive interactions between proteins in the icosahedral viral capsid and host cell receptors to adenovirus-directed innate immunity . The first interaction is mediated by the knob domain of fiber proteins emanating from icosahedral vertices , which bind the coxsackievirus adenovirus receptor ( CAR ) for all serotypes except those belonging to group B [99 , 100] . This is followed by a secondary interaction between a motif in the penton protein located at the base of the vertices and αV integrins [101] . It has been demonstrated that the second interaction activates a PI3K/Akt signaling pathway contributing to NFκB-dependent cytokine expression in cultured epithelial cells [33] . It is also known that membrane rupture and endosomal escape of incoming adenovirus particles activate p38-MAPK and its downstream effector MAPKAP kinase 2 ( MK2 ) , by a mechanism that is dependent on the p38-MAPK kinase MKK6 , but independent of integrin-mediated cell signaling [83] . Similar to other cell stresses , our data indicated that adenovirus cell entry caused EGFR phosphorylation at known p38-MAPK/MK2 substrates ( Ser1046/1047 ) , which have previously been linked to stress-induced EGFR internalization from clathrin-coated pits [86] . The second set of events in the working model describes what is currently known about stress-induced EGFR sorting in MVBs . Previous studies have shown that stress-internalized EGFRs are sorted onto ILVs that have the capacity to back-fuse with the MVB limiting membrane , which facilitates EGFR signaling through cytosolic substrates; and eventually EGFR recycling to the cell surface upon termination of p38-MAPK signaling [44] . Stress-induced EGFR trafficking has also been shown to be regulated by a subset of ESCRT regulatory proteins Hrs , Tsg101 , and Alix [44] . The requirements for Hrs and Alix were not surprising , since both of these ESCRT-associated proteins had already been linked to non-conventional endosomal sorting: Hrs through recognition of hydrophobic amino acid clusters regulating degradation of cytokine receptors [71 , 102]; and Alix by mediating Ub-independent ESCRT-III/MVB sorting of P2Y1 purinergic receptors , and through its known involvement in ILV back-fusion [71 , 102 , 103] However , a role for Tsg101 in sorting non-ubiquitinated cargo such as stress-exposed EGFR remains unclear . The third set of events in the working model summarizes data from this study describing a novel stress-induced EGFR signaling pathway resulting in stabilization and enhanced activity of nuclear NFκB in respiratory epithelial cells . In the canonical NFκB pathway , NFκB subunits are sequestered in the cytoplasm as inactive dimers bound to inhibitory IκB proteins in resting cells [40] . Most upstream stimuli activate NFκB by inducing phosphorylation-dependent proteasomal degradation of IκB proteins [40] . Primarily regulated by inducible IKKs ( inhibitor of NFκB kinases ) , this key step is also catalyzed by Akt , which is known to be activated by capsid engagement of αV integrins [33 , 104] . Liberated NFκB dimers then translocate to the nucleus , where they bind specific DNA sequences via a conserved Rel homology domain ( RHD ) at their N-terminus . One established mechanism for terminating NFκB responses involves newly synthesized IκB proteins induced by activated NFκB , which enter the nucleus , remove NFκB from DNA , and relocalize it to the cytosol [105] . It has also been shown that Pin1 antagonizes negative feedback control of NFκB-p65 signaling by inhibiting binding to IκBα , resulting in increased nuclear accumulation and protein stability of NFκB-p65 and enhanced NFκB activity [54 , 81] . Our results supported a novel hypothesis that cellular stresses , including adenovirus infection and exposure to TNF-α , contributed to sustained activation of NFκB signaling through a non-canonical EGFR pathway associated with phosphorylation of a Thr254-Pro motif in NFkB-p65 , which is a known Pin1 substrate that is unmasked by IκB degradation [54] . Although the EGFR-stimulated pathway regulating proline-directed phosphorylation at Thr254-Pro motif is not currently known , our data supported a possible role for the adaptor protein Gab1 . In addition to sustaining EGFR/ERK-MAPK signaling by facilitating activation of the tyrosine phosphatase Shp2 , full Gab1 activity is known to require trafficking to endosomes [54 , 81 , 106 , 107 , 108] . Gab1 also links EGFRs to multiple signaling pathways including PI3K/Akt signaling cascades , suggesting stress-activated EGFRs may have additional unappreciated roles in viral replication . The fourth set of events in the working model illustrates how RIDα-C2 attenuates stress-induced EGFR/NFκB signaling axis in the absence of other viral proteins . Our studies suggested that RIDα-C2 orchestrated a novel two step process , first by co-opting ESCRT machinery regulating stress-induced EGFR trafficking in MVBs , and then by facilitating lysosomal degradation in the absence of functional Rab7 . In the first mechanism , we showed that the adenoviral RIDα protein attenuated EGFR/NFκB signaling by eliminating the stress-induced Tsg101 trafficking step in MVBs . Tsg101 reportedly controls endosome-to-cytosol release of enveloped viral RNA through an interaction with Alix , implicating a potential role in ILV back-fusion [76] . Since the viral protein formed a molecular complex with Alix , our results supported a novel hypothesis that RIDα blocked back-fusion through competitive inhibition of Tsg101/Alix binding that warrants further investigation . In contrast to the canonical Ub-dependent MVB sorting pathway where Tsg101 was required for degradative EGFR sorting , Tsg101 could have the opposite effect during stress-induced EGFR trafficking by regulating back-fusion and recycling [70] . It will therefore be of interest to determine whether pathological EGFR stress signaling could be down-regulated by manipulating the Tsg101 “steadiness” box , which controls steady-state Tsg101 protein levels under normal physiological conditions [109] . In the second mechanism , RIDα-C2 promoted EGFR degradation . Although MVB sorting silences EGFR signaling by sequestering receptors away from cytosolic effectors , lysosomal degradation appears to be the rate-limiting step in receptor inactivation [110] . Lysosome fusion is regulated in part by RILP , a downstream effector of Rab7 that recruits the HOPS membrane tethering complex to late endosomal compartments via protein-protein interactions with several HOPS subunits [111] . Our prior studies revealed that RIDα-C2 interacted with RILP , and that this interaction was required for adenovirus-induced EGFR down-regulation [77] . We have now shown that Rab7 was dispensable in the adenovirus-induced EGFR trafficking pathway , and that RIDα-C2 expression rescued ligand-induced EGFR degradation following Rab7 gene silencing . However , the RIDα protein encoded by HAdV-B7 failed to reconstitute the ligand-induced pathway , underscoring the potential importance of specific amino acids in the C-terminal tail of RIDα-C2 in regulating EGFR trafficking . Despite Rab7 also having a critical role in autophagosome maturation , however , RIDα-C2 expression did not support starvation-induced autophagy in Rab7-depleted cells [94] . This may reflect the fact that HOPS recruitment during autophagy requires distinct Rab7 effector called PLEKHM1 [112] . The finding that RIDα was a partial Rab7 mimic adds to the growing complexity of molecular mechanisms and biological functions of autophagy during adenovirus infections . On the one hand , we have shown that early adenovirus gene expression and virus production were both enhanced in airway epithelial cells with elevated autophagy , suggesting viral particles were more efficiently released from early endosomes that had fused with autophagosomes [60] . Thus adenovirus and perhaps other respiratory pathogens may co-opt autophagy , which is an important adaptive response to high oxygen pressure in airway epithelia , to help overcome epithelial barrier defenses to infection [113] . Conversely , several early adenoviral transcription units , including E1A , E1B , and E4 , have been implicated in the regulation of autophagic machinery [114 , 115 , 116 , 117] . Adenovirus-induced autophagy is thought to be critical for recycling nutrients that support viral replication , and ultimately for promoting cell lysis necessary for efficient release of new virions [114 , 115 , 116 , 118] . Early adenoviral transcription units act cooperatively , either by activating ( E1A and E1B ) or suppressing ( E4 ) autophagy; and cross-talk between these opposing regulatory mechanisms is thought to prevent excessive autophagy leading to premature cell death before viral replication is complete . The RIDα protein may allow infected cells to calibrate adenovirus-induced autophagic flux to different stresses in the local host microenvironment that activate EGFR signaling in endosomes . RIDα-mediated trafficking could down-tune autophagic flux during the early stages of infection , by competing for a limited cytosolic pool of Rab7 and possibly other rate-limiting shared machinery regulating lysosome fusion during endocytosis and autophagy . In summary , our results supported a working model that RIDα-C2 restored negative feedback control to NFκB signaling , by antagonizing a stress-induced EGFR pathway associated with enhanced NFκB-p65 protein stability and NFκB activity . In addition to RIDα-C2 , two other E3-encoded proteins with seemingly opposing effects on NFκB signaling have been identified: the E3-19K viral protein induced NFκB activity through ER overload [39]; and the E3 protein 14 . 7K inhibited NFκB transcriptional activity through an interaction with the NFκB p50 subunit that blocks DNA binding [119] . Collectively , these E3 proteins may dampen the inflammatory response to group C adenoviruses , by controlling NFκB-driven release of the neutrophil chemoattractant IL-8 from infected epithelial cells [120 , 121] . Alternatively , E3 proteins could fine-tune combinatorial control of NFκB signaling , which seems to be important for protecting epithelial cells from inflammation caused by innate immune cells recruited to epithelial surfaces of infected cells [32] . Our studies supported future efforts to determine whether serotype-specific differences in RIDα and other E3 proteins contribute to inflammatory disease associated with HAdV-B7 , which is known to induce IL-8 protein production by a mechanism requiring virus internalization but not viral protein expression [122] . Our results also suggested RIDα-C2 could limit the innate immune response to E1A-deleted adenovirus vectors . There is increasing evidence that many viruses exploit EGFR function to facilitate their replication and antagonize host antiviral responses [123] . Until now it was generally assumed that viruses co-opted mechanisms induced by ligand-receptor interactions . Recognition that adenovirus contributed to NFκB signaling by activating a non-canonical EGFR pathway is significant because unique host proteins regulating this pathway represent novel drug targets for therapeutic development .
The RIDα antibody was commercially produced in rabbits using a peptide corresponding to the C-terminal 15 amino acids in the protein encoded by the HAdV-C2 serotype ( Rockland Antibodies and Assays; Limerick , PA ) [49] . See Table 1 for a comprehensive list of primary antibodies used in this study . All secondary antibodies were purchased from Jackson ImmunoResearch ( West Grove , PA ) . Adenocarcinomic human alveolar basal epithelial A549 cells purchased from ATCC ( catalog number CCL-185 ) were authenticated utilizing Short Tandem Repeat ( STR ) profiling . A549 cells with stable expression of a FLAG-tagged RIDα-C2 gene from HAdV-C2 are described in [89] . The RIDα protein from HAdV-B7 ( RIDα-B7 ) cloned in pcDNAI/Amp was used as template for a PCR reaction using forward ( 5’-ATCGTAAAGATCTTGATTCCTCGAGTTCTTATATTATTG-3’ ) and reverse ( 5’-CTAAGATCTCCTTAAAGAATTCTGAGAAGATCAGCTATAGTCCTG-3’ ) primers to amplify the RIDα-B7 open reading frame , and incorporate flanking BglII restriction sites ( underlined ) . PCR products were digested with BglII , and then ligated to an amino-terminal FLAG epitope in the polylinker region of the pExchange2 plasmid ( Stratagene , La Jolla , CA ) digested with the same restriction enzyme . GP2-293 retrovirus packaging cells were transfected with the FLAG-tagged RIDα-B7 encoding plasmid using Trans-IT 293 transfection reagent ( Mirus Bio ) . Pantropic retrovirus was generated upon subsequent transfection of drug-selected packaging cells with pVSV-G plasmid . Retrovirus-containing media was collected 48 h later and added to A549 cells , followed by G418 selection . Stable RIDα-B7 expression was verified by immunoblotting and immunostaining with FLAG antibodies . A549 cells and their derivatives were maintained in Ham’s F12 medium supplemented with 10% fetal bovine serum and 2 mM glutamine . Mouse NIH3T3 fibroblasts ( established from NIH Swiss mouse embryo ) expressing WT human EGFR , or a kinase-dead EGFR construct with a K721M mutation that abrogates ATP binding ( gift of Axel Ulrich; Max Planck Institute of Biochemistry ) , were maintained in Dulbecco’s-modified MEM medium supplemented with 10% FBS and 2 mM glutamine [19 , 51] . CHO ( Chinese hamster ovary ) cells ( gift of Martin Snider , Case Western Reserve University ) were grown in MEM-alpha medium supplemented with 10% FBS and 2 mM glutamine . Human adenoviruses HAdV-C2 , HAdV-B7 , and HAdV-E4 were purchased from ATCC ( catalog numbers VR-846 , VR-7 and VR-1572 respectively ) . A RIDα-null HAdV-C2 mutant virus that deleted 107 base pairs in the amino terminal region of the viral protein was described in [65] . Adenoviruses were propagated , and multiplicity of infection or MOI was determined by plaque assay in A549 cells , according to standard methods [124] . It has been estimated that an MOI of at least 5 to 10 is required to ensure that 100% of cells are infected in tissue culture [125] . In preliminary studies , it was established that an MOI of 50 to 100 triggered EGFR stress responses prior to the onset of viral gene transcription , and sufficient RIDα expression to counter these responses , in human A549 cells; and that an MOI of 200 to 250 induced these responses in mouse fibroblasts . The GFP-tagged Vps22 mammalian expression plasmid was a gift from Dr . Cecilia Bucci ( Università di Lecce ) [126] . Gene silencing studies were carried out using ON-TARGETplus Human siRNA Smart Pools from Dharmacon ( Lafayette , CO ) listed in Table 2 , which were introduced to A549 cells using Oligofectamine Reagent exactly as described in [60] . Cells were metabolically labeled with 35S-Express Protein Labeling Mix ( 2 . 5 mCi/ml; PerkinElmer Life Sciences , Boston , MA ) diluted in methionine and cysteine-free medium . Radiolabeled cells were rinsed 3 times with PBS supplemented with 5 mM EDTA , 5 mM EGTA , and a phosphatase inhibitor cocktail ( 10 mM NaF , 10 mM Na4P2O7 , and 1 mM Na3VO4 ) ( PBS+ ) ; and lysed with 1% NP-40 in a solution of 50 mM Tris-HCl , pH 7 . 5 , supplemented with 150 mM NaCl , the phosphatase inhibitor cocktail , and a cocktail of protease inhibitors ( 100 μM phenylmethylsulfonyl fluoride , 10 μg/ml aprotinin , 10 μg/ml leupeptin , 4 μg/ml pepstatin ) ( IP lysis buffer ) . Immune complexes were recovered by incubating lysates with antibodies of interest at 4°C overnight followed by 1-h incubation with protein A conjugated to Sepharose CL-4B beads , followed by extensive washing with IP lysis buffer . Radioactive proteins were eluted and resolved by SDS-PAGE for detection by fluorography using standard techniques . For Alix co-IP studies , cells were harvested by trypsinization , washed 2 times with PBS+ , and re-suspended in 10 mM HEPES , pH 7 . 4 , supplemented with 142 . 5 mM KCl , 0 . 2% NP-40 , 2 mM NaVO4 , 20 mM NaF , 10 mM CuCl2 , and the protease inhibitor cocktail ( co-IP lysis buffer ) . Cells were homogenized with 10 strokes in a Dounce homogenizer , and clarified lysates were incubated with a biotin-conjugated FLAG antibody for 2 h at 4°C . Lysates were then incubated with streptavidin beads overnight at 4°C , and immune complexes were washed 3 times with co-IP lysis buffer , eluted with Laemmli buffer , and resolved by SDS-PAGE to detect protein complexes by immunoblotting . Equal protein aliquots ( determined by Bradford assay ) of total cell lysates , or nuclear and cytoplasmic fractions prepared using a kit from Cell Biolabs ( catalog number AKR-171 ) , were resolved by SDS-PAGE for immunoblot analysis . Nitrocellulose filters were incubated with primary antibodies and appropriate HRP-conjugated secondary antibodies diluted in blocking solution supplemented with 5% evaporated dry milk , for detection by enhanced chemiluminescence ( Amersham Life Sciences ) . Protein bands of interest and background measurements below each protein band , which were deducted from protein band values , were quantified with the ImageJ image processing program from the National Institutes of Health . Background-subtracted EGFR protein bands were normalized to background-subtracted loading controls for each sample , and data are presented as fold-change relative to the band at time 0 for each siRNA treatment . GST fusion proteins with RIDα cytoplasmic tail peptide fragments were described previously in [52] . Fusion proteins were purified from BL21 competent E . coli using the Inclusion Body Solubilization Reagent from Thermo Scientific ( catalog number 78115 ) , according to the manufacturer’s instructions . Inclusion body lysates were incubated with glutathione-Sepharose beads ( Amersham-Pharmacia , catalog number 17075601 ) overnight at 4°C with rotation followed by three washes with a solution of 50 mM Tris ( pH 7 . 4 ) , 10 mM MgCl2 , 0 . 15 M NaCl , and 1% Triton X-100 . Beads with attached fusion proteins were incubated with whole cell lysates prepared from CHO cells using IP buffer . Beads were washed four times with IP lysis buffer , solubilized with sample buffer , resolved by SDS-PAGE , and immunoblotted with Alix antibody to detect bound proteins . Cells were seeded on glass cover-slips coated with poly-L-lysine ( Sigma-Aldrich , catalog number P4707 ) for confocal imaging . Cells were perforated with 0 . 5% β-escin , which is a naturally derived saponin mixture , diluted with a solution of 80 mM PIPES , pH 6 . 8 , supplemented with 5 mM EGTA and 1 mM MgCl2 for 5 min and fixed with 3% paraformaldehyde–PBS for 15 min as described previously [51] . Non-specific binding was blocked with 5% normal serum from the host animal used to generate the secondary antibody ( Jackson ImmunoResearch Laboratories; West Grove , PA ) . Cells were stained with primary or secondary antibodies overnight at 4°C or 1 h at room temperature . Antibodies were diluted in PBS supplemented with 0 . 5% β-escin and 3% radioimmunoassay-grade BSA . Single vertical confocal images were acquired with a Zeiss LSM 510 Meta laser scanning microscope ( Carl Zeiss MicroImaging , Jenna , Germany ) using diode ( excitation 405 nm ) , Argon ( excitation 488 nm ) , and HeNe ( excitation 543 and 633 nm ) lasers , 40× or 100× Plan Apo NA 1 . 4 objectives , and Zeiss LSM software ( Carl Zeiss MicroImaging , Jenna , Germany ) . Confocal images were overlayed with phase contrast images to draw cell and nucleus outlines using graphics software . Quantification of co-localization was performed by measurement of Mander's coefficient in at least 10 cells per experiment using ImageJ . Infected cells were rinsed once with chilled serum-free Dulbecco's modified MEM , incubated with a monoclonal antibody EGFR1 directed towards an external EGFR conjugated to 10-nm colloidal gold particles ( Electron Microscopy Sciences ) by the tannic acid procedure for 1 h and then re-cultured at 37°C for 30 min [19 , 127] . Cells were fixed with a solution of 2 . 5% glutaraldehyde , 2% formaldehyde , and 0 . 1 M sodium cacodylate , pH 7 . 4 , for 15 min on ice , rinsed with 0 . 1 M sodium cacodylate , pH 7 . 4 , and post-fixed with 1% osmium tetroxide for 1 h on ice . Fixed cells were treated overnight at 4°C with 1% uranyl acetate , rinsed three times with water , dehydrated with ethanol , and embedded in polybed resin for 3 days before being baked at 60°C . Thin sections ( 80-nm ) were mounted on Formvar nickel-coated grids , and cells were counterstained with Reynold's lead citrate and 2% uranyl acetate and then examined on a JEOL 100 CX electron microscope . A total of 50 cells were examined for distribution of gold particles located on MVB limiting membrane or on ILVs , in two independent experiments . The means of the total gold particles localized in MVB compartments were compared using a standard two-tailed Student's t test . Tissue culture supernatants were collected from A549 cells following various treatments , and stored at -80°C . IL-8 ELISA was performed using an anti-IL-8 neutralizing monoclonal primary antibody , biotinylated anti-IL-8 polyclonal secondary antibody and recombinant human IL-8 protein standards ( Boster Biological Technology , catalog number EK0413 ) . The plates were developed using avidin-horseradish peroxidase conjugate and TMB substrate , and absorbance read at 450 nm using a BioTek Synergy HT instrument . Statistical analyses were performed using the Student’s t test . P-values < 0 . 05 were considered to be statistically significant . Computer-generated images were minimally processed , and all processing was applied equally to all parts of each image as well as controls , using Adobe Photoshop CS5 . 1 software package . Figures were prepared with Adobe Illustrator CS5 . 1 software package .
|
Although most adenovirus infections produce a mild and self-limiting disease , they can be life threatening for immunocompromised individuals . Some serotypes also cause epidemic outbreaks that pose a significant health risk in people with no known predisposing conditions . Although the early region 3 ( E3 ) of the adenovirus genome is known to play a critical role in viral pathogenesis , experimental evidence regarding the molecular mechanisms effecting damage in the host is still limited . Here we provide the first studies showing that adenovirus infection induced stress-activated EGF receptor ( EGFR ) pro-inflammatory signaling prior to nuclear translocation and transcription of viral DNA in non-immune epithelial target cells . We have also identified host molecular mechanisms co-opted by the E3 RIDα protein that potentially limit immune-mediated tissue injury caused by stress-activated EGFR . There is increasing evidence that many viruses exploit EGFR function to facilitate their replication and antagonize host antiviral responses . Until now , it was generally assumed that viruses co-opted mechanisms induced by conventional ligand-regulated pathways . Recognition that stress-activated EGFR signaling may play a critical role in viral pathogenesis is significant because unique host proteins regulating this pathway represent novel drug targets for therapeutic development .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"protein",
"transport",
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"cellular",
"stress",
"responses",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"probe",
"techniques",
"pathogens",
"cell",
"processes",
"immunology",
"microbiology",
"egfr",
"signaling",
"immunoblotting",
"viruses",
"dna",
"viruses",
"molecular",
"biology",
"techniques",
"antibodies",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"stress",
"signaling",
"cascade",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"post-translational",
"modification",
"viral",
"pathogens",
"physiology",
"adenoviruses",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"organisms",
"signaling",
"cascades"
] |
2019
|
Adenovirus early region 3 RIDα protein limits NFκB signaling through stress-activated EGF receptors
|
Evasion of immune T cell responses is crucial for viruses to establish persistence in the infected host . Immune evasion mechanisms of Epstein-Barr virus ( EBV ) in the context of MHC-I antigen presentation have been well studied . In contrast , viral interference with MHC-II antigen presentation is less well understood , not only for EBV but also for other persistent viruses . Here we show that the EBV encoded BZLF1 can interfere with recognition by immune CD4+ effector T cells . This impaired T cell recognition occurred in the absence of a reduction in the expression of surface MHC-II , but correlated with a marked downregulation of surface CD74 on the target cells . Furthermore , impaired CD4+ T cell recognition was also observed with target cells where CD74 expression was downregulated by shRNA-mediated inhibition . BZLF1 downregulated surface CD74 via a post-transcriptional mechanism distinct from its previously reported effect on the CIITA promoter . In addition to being a chaperone for MHC-II αβ dimers , CD74 also functions as a surface receptor for macrophage Migration Inhibitory Factor and enhances cell survival through transcriptional upregulation of Bcl-2 family members . The immune-evasion function of BZLF1 therefore comes at a cost of induced toxicity . However , during EBV lytic cycle induced by BZLF1 expression , this toxicity can be overcome by expression of the vBcl-2 , BHRF1 , at an early stage of lytic infection . We conclude that by inhibiting apoptosis , the vBcl-2 not only maintains cell viability to allow sufficient time for synthesis and accumulation of infectious virus progeny , but also enables BZLF1 to effect its immune evasion function .
Successful persistence of viral infection depends on the establishment of a balance between host immune responses and viral immune evasion . Epstein-Barr virus ( EBV ) , which is carried by more than 90% of the adult human population worldwide , is a prime example of a persistent virus that is generally harmless , but which can cause serious disease including various tumours [1] . A number of immunoevasins of EBV have recently been identified as acting at different points along the MHC class I ( MHC-I ) presentation pathway to modulate recognition by CD8+ T cell responses [2]–[7] . However , with regards to evasion of the MHC class II ( MHC-II ) antigen presentation pathway , EBV is less well understood . Virus-specific CD4+ T cell responses , which include some clones with cytotoxic activity , are broadly distributed against numerous proteins encoded by the EBV genome; both latent protein antigens [8] and the larger number of lytic protein antigens [9] , [10] . These observations indicate a need for EBV to also modulate MHC-II antigen presentation pathways , particularly during in EBV lytic cycle . Mechanisms for interfering with the MHC-II antigen presentation pathway have been reported for other herpesvirus; for example , US2 , US3 and pp65 of cytomegalovirus [11]–[13] and glycoprotein B ( gB ) of herpesimplex virus type 1 [14] . However , EBV has no homologues to these immune evasion genes of CMV , and there is no evidence that the homolog of HSV-1 gB protein encoded by the BALF4 gene of EBV targets the MHC-II pathway . An unrelated EBV glycoprotein , gp42 , has however been shown to associate with MHC-II molecules and to inhibit antigen presentation to CD4+ T cells [15] , [16] . More recently , the immediate-early EBV gene BZLF1 , which encodes a transcription factor initiating EBV lytic cycle , was reported to be a potential modulator of MHC-II antigen presentation [17] . Ectopic expression of BZLF1 in Raji cells inhibited the expression of MHC-II molecules , apparently through repression of CIITA transcription . However , interpretation of these data is complicated by the fact that Raji is an EBV-carrying B cell line in which expression of BZLF1 can initiate virus lytic cycle and , therefore , the expression of other viral genes that may be responsible for modulating MHC-II expression . One such candidate is the early antigen , BGLF5 , which has been shown to induce global mRNA degradation and thereby to reduce expression of various host proteins , including MHC-I and MHC-II [2] , [3] . In addition , as the level of expression of MHC molecules does not necessarily reflect the degree of T cell recognition , it is important to assay antigen presentation using functional T cell recognition . In this study , we demonstrated for the first time that BZLF1 does indeed impair recognition by EBV-specific CD4+ T cells . However , this was not the result of downregulation of MHC-II α/β chain molecules as expected , but through downregulation of the invariant chain ( Ii , or CD74 ) which serves as a chaperone to ensure correct loading of antigenic peptide fragments . As CD74 also mediates cell survival , its downregulation by BZLF1 induced cell death . A pivotal role was also identified for a vBcl-2 ( BHRF1 ) during virus lytic cycle . By abrogating the concomitant toxic properties of BZLF1 , the vBcl-2 enabled the immune evasion function of BZLF1 .
To gain a complete picture of effects on the antigen presentation pathway , it is crucial to include the functional T cell recognition assays as readout . We sought to determine whether BZLF1 can impair the MHC-II antigen presentation of newly processed MHC-II/peptide complexes and also of pre-existing MHC-II/peptide complexes . Recognition of newly processed MHC-II/peptide complexes was assayed by pulsing target cells with virus particles , before co-culturing with BXLF2-specific ‘LEK’ CD4+ effector T cells recognising a virion antigen . Pre-existing MHC-II/peptide complexes were assayed with EBNA1-specific ‘SNP’ CD4+ effector T cells to detect EBNA1 antigen that is expressed in both the latent and lytic phases of infection in EBV-transformed B lymphoblastoid cell lines ( LCL ) . Whilst antigen processing for presentation via MHC-II complexes is generally considered to occur mainly through endocytosis of exogenous antigen ( e . g . virus particles ) , it is also possible for endogenously expressed antigen to be processed by directly accessing the MHC-II antigen processing pathway [18] , [19] . EBNA1 is paradigmatic for endogenous processing of target peptides for recognition by immune CD4+ T cells [18] , [20] . In our initial experiments we generated LCL targets with a BZLF1-knockout recombinant EBV ( BZLF1KO-LCLs ) , and transfected them with a doxycycline ( DOX ) -inducible BZLF1-expressing vector , pRTS-CD2-BZLF1 . This vector expresses a CD2 surface marker from the vector backbone , allowing enrichment of viable transfected cells at 24h after transfection . The BZLF1 gene and an NGFR-IRES-GFP sequence were regulated by a bi-directional DOX-regulated promoter , which allows a further round of purification of BZLF1-expressing cells following DOX-induction . Control BZLF1KO-LCL targets were transfected with a vector containing a non-sense reverse BZLF1 sequence insert . Details of the plasmid vectors and the isolation of transfected cells are described in Figure 1A . BZLF1- and control-vector transfected LCLs were treated with DOX for 24 h before separating the induced population according to NGFR expression . From each induced line , two populations were obtained: NGFR–/GFP– cells lacking transfected plasmid , and NGFR+/GFP+ cells in which BZLF1/GFP ( or control/GFP ) expression was evident ( Figure 1B ) . The different populations were used as target cells in CD4+ T cell assays where recognition was measured by assaying the release of IFN-γ from the effector T cells . GFP– cells from both the pRTS-CD2-control and the pRTS-CD2-BZLF1 transfectants were equally well recognised , both by BXLF2-specific CD4+ effectors following uptake and processing of virions ( Figure 1C , upper histogram ) and by EBNA1-specific CD4+ effectors ( Figure 1D; upper histograms ) . In contrast , recognition of newly processed BXLF2 peptides from virus-pulsed targets ( Figure 1C , lower histogram ) and pre-existing peptides from EBNA1 ( Figure 1D , lower histogram ) was impaired in GFP+ cells expressing BZLF1 protein . The results illustrated in Figs . 1C and 1D are from one representative experiment . In three independent experiments , recognition of newly processed BXLF2 antigen by ‘LEK’ CD4+ T cells was reduced by 40–60% , and recognition of pre-existing EBNA1 peptide by ‘SNP’ CD4+ T cells was reduced by 25–40% following expression of BZLF1 . These experiments demonstrate that BZLF1 expression in LCLs leads to impaired EBV-specific CD4+ effector T cell recognition . In the BXLF2-specific CD4+ T cell recognition assays , there was very low but consistent recognition of BZLF1-expressing LCLs that had not been pulsed with EBV particles ( bottom histogram , Figure 1C ) . This suggested that even within the short time-frame of this experiment the expression of BZLF1 can trigger lytic cycle and produce small but sufficient amounts of virus antigen for MHC-II processing . This conclusion was supported by further assays where expression of BZLF1 led to clear recognition not only by BZLF1-specific CD4+ effectors , but also by CD4+ effectors specific for the BMRF1 early antigen or the gp350 late antigen ( Figure S1 ) . Immunoblot analysis of the LCL targets showed that induction of BZLF1 led to expression of EBV lytic proteins ( Figure 1E ) . Thus , whilst the data in Figure 1 are consistent with BZLF1 interfering with CD4+ T cell responses , further experiments with EBV-negative targets were necessary to avoid complications arising from BZLF1 initiation of the cascade of lytic cycle expression . To investigate the effect of BZLF1 on MHC-II antigen presentation in EBV-negative target cells , we developed a strategy that allowed analysis of transiently transfected target cells . As intercellular antigen transfer is never detectable for EBNA1 [20] , we chose this as the target antigen . To increase the sensitivity of the assay , a mutant form of EBNA1 ( Cyto-EBNA1 ) was used as its cytosolic location renders its processing to MHC-II molecules more efficient than wild-type nuclear EBNA1 [20] . EBV-negative MJS melanoma cells expressing MHC-II DRB5*01 were transfected with cyto-EBNA1 expression plasmid together with IRES-GFP plasmid vectors for BZLF1 or other EBV lytic genes . The cultures were then assayed for antigen presentation to MHC-II DRB5*01-restricted CD4+ T cells specific for the ‘SNP’ EBNA1-derived peptide . T cell recognition was measured by the release of IFN-γ from the effector cells . The results of 5 independent experiments are summarised in Figure 2A . Mock transfected cells were not recognised , but good recognition was seen with target cells co-transfected with cyto-EBNA1 and control IRES-GFP vector . Recognition of cyto-EBNA1 was not significantly affected by co-transfection of BZLF2 or BALF4 vectors expressing the EBV lytic proteins gp42 or gp110 , respectively . However , recognition of cyto-EBNA1 was substantially and reproducibly reduced by 60–80% when the target cells were co-transfected with BZLF1 . Replicate aliquots of cells analyzed by immuno-blots confirmed expression of BZLF1 in the cells transfected with the BZLF1-GFP plasmid ( Figure 2B ) . There was no obvious change of MHC-II DR expression in any of the target cells . Whilst the levels of EBNA1 were similar in the cells co-transfected with cyto-EBNA1 and IRES-GFP , BZLF2-GFP or BALF4-GFP plasmids , there was an unexpected and marked reduction of EBNA1 expression when co-transfected with the BZLF1-GFP plasmid ( Figure 2B ) . Two colour flow cytometry analysis of viable cells stained for surface MHC-II DR suggested that none of the EBV lytic genes tested reduced the level of cell surface DR molecules on GFP+ cells compared to the expression on untransfected GFP– cells in the same cultures ( Figure 2C ) . Notably , there were relatively few GFP+ cells in the cultures transfected with BZLF1-GFP; typically less than 1% compared with 15% in cultures transfected with control IRES-GFP plasmid . This indicated a toxic effect of BZLF1 in these cells , which could account for the reduced EBNA1 target antigen expression ( Figure 2B ) and also the reduced T cell recognition ( Figure 2A ) . Toxicity of BZLF1 was confirmed in other cell types , including an EBV-negative subclone of the Akata Burkitt's lymphoma , Akata-A3 . Here , we used the DOX-inducible pRTS-CD2-BZLF1 as in Figure 1 . Akata-A3 cultures transfected with either pRTS-CD2-BZLF1 or pRTS-CD2-control plasmids were enriched by CD2 selection to around 80% purity . Following treatment with DOX , the percentage of GFP+ cells was monitored over a period of 72 h . For the first 24 h of DOX treatment , both control- and BZLF1-transfected cultures contained indistinguishable numbers of induced GFP+ cells , but thereafter the BZLF1 transfected cultures showed a drop in the percentage of GFP+ cells such that by 72 h these BZLF1 cultures contained only around half the percentage of GFP+ cells as the control cultures ( Figure 3A ) . These results indicate that BZLF1 is toxic to Akata-A3 , but with slower kinetics than in MJS cells . The slower kinetics of toxicity in Akata-A3 allowed examination of cellular protein expression at 48 h after BZLF1 induction . Pure populations of GFP+ cells were obtained by flow cytometer sorting for GFP expression , and were analysed by immunoblotting . The first important point to note is that physiological levels of BZLF1 were obtained in these cells . Titration of the A3-BZLF1 transfectants against an LCL containing 5% BZLF1+ cells ( i . e . in lytic cycle ) , followed by densitometry analysis of immunoblots , indicated that the level of BZLF1 in the transfected cells was around 115% of that in lytically-infected LCLs ( Figure 3B ) . Secondly blotting for Bcl-2 and Bcl-xL expression showed a clear downregulation of these anti-apoptotic proteins in the BZLF1-expressing Akata-A3 cells ( Figure 3C ) , which correlated with a downregulation of Bcl-2 and Bcl-xl mRNA transcripts ( Figure 3D ) . EBV encodes a well-characterised Bcl-2 homolog , BHRF1 , as an early lytic cycle protein [21] . We therefore investigated whether the toxicity of BZLF1 could be reversed by this vBcl-2 . To address this possibility , we first examined the MJS line as it appears to be particularly sensitive to BZLF1 . In the representative experiment shown in Figure 4A , MJS cells were transfected with IRES-GFP , BZLF1-GFP , or with BZLF1-GFP and a BHRF1 expression plasmid . Cultures were sampled at various time points post-transfection for analysis of GFP expression by flow cytometry . In the IRES-GFP control transfectants , about 10% of the viable cells were GFP+ at 8 h , rising to around 17% by 30 h . In marked contrast , MJS cells transfected with BZLF1-GFP showed only about 4% GFP+ viable cells at 8 h post-transfection , and thereafter the percentage of GFP+ cells gradually fell at later time points . However , when a BHRF1 expression vector was co-transfected with the BZLF1-GFP plasmid , the percentage of viable GFP+ cells continued to increase to about 10% by 30 h . As in the previous set of experiments with Akata-A3 transfectants , the toxicity in MJS cells was mediated by physiological levels of BZLF1 ( Figure 4B ) . Interestingly , the partial reversal of toxicity by co-expressed BHRF1 was achieved by levels of the vBcl-2 that were substantially lower than the levels observed in lytically-infected LCLs ( Figure 4B ) . BHRF1-mediated protection from BZLF1-induced toxicity was also observed in other EBV-negative B cells ( Figure S2 ) . These results suggest that both the toxicity of BZLF1 and its reversal by BHRF1 are phenomena that are likely to be physiologically relevant during normal lytic cycle in EBV-infected cells . The protection afforded by BHRF1 in MJS cells enabled us to re-examine the effect of BZLF1 on MHC-II antigen presentation without the confounding effect of induced cell death . MJS cells co-transfected with BHRF1 together with either IRES-GFP or BZLF1-GFP were examined for expression of cellular proteins in viable GFP+ sorted subpopulations at 48 hr post-transfection . Immuno-blot analysis of cell lysates showed that , as with the Akata-A3 cells ( Figure 3C ) , BZLF1 also downregulated Bcl-2 and Bcl-xl protein expression in MJS cells ( Figure 4C ) . Interestingly , whilst the total cellular level of MHC-II DRα molecules was only slightly reduced following BZLF1 expression , the level of invariant chain , CD74 , was markedly downregulated ( Figure 4C ) . In BZLF1-transfected cultures , cell surface MHC-II DR expression was reproducible slightly elevated by around 10–20% in the GFP+ gated cells ( Figure 4D , top right histogram ) ; in contrast , there was a marked reduction in the expression of CD74 ( about 50% ) on the surface of GFP+ cells compared with GFP– cells ( Figure 4D , lower right panel ) . One possible explanation for the discordance between total and cell surface MHC-II DR levels may be an altered localisation of MHC-II molecules when CD74 is downregulated . Analysis of control transfections showed that the levels of cell surface MHC-II DR ( Fig 4D , top left histogram ) and CD74 ( bottom left histogram ) were indistinguishable between GFP+ and GFP– gated cells in the same culture . Similar experiments performed with the Akata-A3 cell line using the DOX inducible vector ( Fig 5A ) revealed that expression of BZLF1 in these B cells gave similar results to those seen with MJS cells . To study the surface levels of MHC-II DR and CD74 in the context of lytic cycle in normal EBV-transformed normal B cells , we examined selected LCL cultures that showed a clear subpopulation of cells spontaneously in lytic cycle . These LCLs were stained for surface DR or CD74 on viable cells , then were fixed and permeabilized for staining of intracellular BZLF1 . The representative result in Figure 5B shows that the BZLF1+ cells spontaneously entering lytic cycle also showed a slight elevation of surface MHC-II DR ( around 20% ) and a clear reduction of surface CD74 ( typically , 30–40% ) compared with the BZLF1– latently infected population in the same LCL culture , i . e . mirroring the results obtained with BZLF1-transfected cells in previous experiments . Interestingly , when these same LCL cultures were co-stained for VCA to analyse the minor subpopulation of lytically-infected cells that have progressed to late lytic cycle , both MHC-II DR and CD74 were seen to be downregulated ( Figure 5C ) . As both MHC-II and CD74 can be transcriptionally regulated by CIITA [22] , and BZLF1 was previously reported to transcriptionally repress the CIITA promoter [17] , we investigated whether BZLF1 in our experiments might be modulating CD74 transcription via CIITA . Using HEK293 as a model CIITA–/MHC-II–/CD74– line , we overexpressed CIITA from a heterologous promoter , which in turn induced MHC-II and CD74 protein expression . However , when BZLF1 was co-expressed , surface expression of MHC-II DR was slightly elevated and CD74 was markedly downregulated ( Figure S3 ) , exactly as seen in our earlier experiments with MJS and B cells . This suggests that BZLF1 is modulating CD74 expression by a CIITA-independent mechanism . Furthermore , in Akata-A3 B cells , whilst MHC-II DR transcripts were slightly reduced following BZLF1 expression , no reduction in CD74 transcripts was observed ( Fig 6A ) . Pulse-chase metabolic labeling experiments with 35S-methionine , revealed that BZLF1 has no effect on CD74 translation or protein maturation ( Figure 6B ) . The exact mechanism of CD74 regulation was not actively pursued further , although experiments performed for other reasons ( e . g . Figure S4 ) indicated that BZLF1 may be regulating trafficking of CD74 to or from the plasma membrane . We next revisited whether BZLF1 retains the ability to impair CD4+ T cell recognition when its toxicity is attenuated by BHRF1 . MJS cells were co-transfected with cyto-EBNA1 target antigen plasmid and control-IRES-GFP or BZLF1-GFP plasmid vectors without ( Figure 7A ) or with ( Figure 7B ) BHRF1 expression plasmid . As in earlier experiments , immune CD4+ T cell recognition of the processed EBNA1 target was substantially impaired by expression of BZLF1 in the absence of BHRF1 ( Figure 7A , histogram ) , which correlated with a clear reduction in the amount of EBNA1 antigen expression ( Figure 7A , blots ) . However , when co-expressed with BHRF1 the BZLF1 had no significant effect on the levels of EBNA1 target antigen ( Figure 7B , blots ) , and the immune CD4+ T cell recognition was inhibited to an even greater extent than when BHRF1 was not expressed ( Figure 7A , histogram ) . These data demonstrate that BZLF1 can indeed interfere with MHC-II antigen presentation to cause a substantial impairment of CD4+ effector T cell recognition . As BZLF1 impairs T cell recognition without downregulating MHC-II DR expression , we considered it likely that the downregulation of CD74 might be responsible by qualitatively altering the MHC-II/peptide complexes available at the cell surface . To directly test this , we first generated an MJS line in which CD74 was over-expressed from a strong heterologous promoter , to see if we could maintain high levels of surface CD74 after BZLF1 expression and reverse the impaired CD4+ T cell recognition . However , despite massive over-expression of total cellular CD74 , the amount of CD74 at the cell surface was barely increased; and the ability of BZLF1 to downregulate surface CD74 was unaffected ( Figure S4 ) . This is consistent with BZLF1 downregulating surface CD74 by a post-translational trafficking mechanism , but the experiment was otherwise uninformative . We next carried out the reverse experiment , asking whether downregulation of CD74 by itself was sufficient to impair CD4+ T cell recognition . Again , we first tested this in MJS cells , using an shRNA inhibition approach . Successful knockdown of CD74 in MJS cells was associated with considerable cell death . However , using MJS-BHRF1 cells , we achieved efficient knockdown of CD74 and retained cell viability ( Figure 8A and 8C ) . Importantly , knockdown of CD74 was associated with significant impairment of CD4+ T cell recognition of transiently expressed cyto-EBNA1 ( Figure 8B , upper histogram ) . This was a specific effect , as pre-incubation of control and CD74-KO targets with saturating concentrations of synthetic EBNA1 target peptide resulted in both targets being equally-well recognised ( Figure 8B , lower histogram ) . Immunoblots confirmed that the downregulation of CD74 in this experiment had no effect on the expression of the EBNA1 target antigen ( Figure 8C ) . Finally , shRNA-mediated knockdown of CD74 in EBV-transformed LCLs ( Figure 8D ) resulted in a similarly impaired recognition by CD4+ effector T cells ( Figure 8E , left histogram ) . The specificity of this effect in LCLs was demonstrated by the observation that CD74 knockdown had no effect on CD8+ effector T cell recognition ( Figure 8E , right histogram ) .
We have demonstrated that the EBV-encoded BZLF1 protein can interfere with recognition by immune CD4+ effector T cells , but it comes at a cost of toxicity . During EBV lytic cycle , this toxic property of BZLF1 can be overcome by expression of the vBcl-2 BHRF1 at an early stage of lytic infection . Unexpectedly , the impaired CD4+ effector T cell recognition did not correlate with levels of surface MHC-II molecules , but rather with a marked downregulation of CD74 ( Ii ) on the surface of target cells . CD74 not only facilitates appropriate peptide loading to MHC-II complexes in the endolysosomal vesicles [23]–[25] but , as a surface receptor for Macrophage Migration Inhibitory Factor ( MIF ) , it also enhances cell survival through transcriptional upregulation of Bcl-2 family members [26] , [27] . Therefore , downregulation of CD74 is a likely mechanism accounting for both the immune-evasion and death-inducing functions of BZLF1 . CD74 and MHC-II genes can both be regulated by the transcription factor , CIITA [22] , and in many instances CD74 and MHC-II genes appear to be co-ordinately expressed . As CIITA expression is negatively regulated by BZLF1 through the suppression of its promoter [17] , we might have expected that BZLF1 would downregulate both CD74 and MHC-II . However , from the present work , it is clear that BZLF1 selectively downregulates CD74 . Transcription of CD74 is not solely regulated by CIITA [28] , which could potentially explain discordant regulation of CD74 and MHC-II genes . Nevertheless , we found that the dominant mechanism by which BZLF1 downregulates CD74 was in fact post-translational . Although BZLF1 can inhibit CIITA transcription [17] , we consistently found cell surface MHC-II DR to be slightly elevated in BZLF1 expressing EBV-negative cells ( Figure 4D , 5A ) . This mirrors what is observed following synchronous induction of lytic cycle in EBV-positive B cells , where cell surface MHC-II DR is initially elevated although it then falls between 12 to 24 h post-induction to a level that is around 40% of that in latently infected cells [29] . In the present study , we also observed that the MHC-II DR is slightly elevated in BZLF1+ cells in spontaneously lytic LCLs ( Figure 5B ) , but that the level of surface DR was reduced in the minor subpopulation of cells expressing late viral capsid antigen ( Figure 5C ) . Together , these results suggest that the initial rise in surface MHC-II DR expression in lytically infected cells is likely to be due to BZLF1 expression , while the later reduction in MHC-II DR in lytic cycle may be due to BGLF5 , which acts as a host shutoff protein and contributes to immune evasion [2] , [3] , and/or a delayed effect of BZLF1-mediated inhibition of CIITA . CD74 is a polypeptide involved in the transport and peptide loading of MHC-II molecules [23]–[25] , [30] . Newly synthesized MHC-II α and β chains complex with CD74 ( invariant chain ) in the endoplasmic reticulum . A cytosolic di-leucine-targeting motif of CD74 directs MHC-II complexes to the endocytic pathway , either directly from the trans-Golgi network or via rapid internalization from the cell surface . The majority of CD74 at the cell surface is physically associated with MHC-II molecules [31] and most , if not all , of immature MHC-II molecules ( complex of α chain , β chain and invariant chain ) reach the cell surface before entering the peptide-loading compartment [32] . CD74 is rapidly turned over at the cell surface . Downregulation of surface CD74 by BZLF1 may therefore indicate a reduction of available immature MHC-II complexes for processing and uptake of antigenic peptides in the endosomes , and would account for the marked effect of BZLF1 expression on the MHC-II antigen processing pathway . Indeed , when we targeted expression of CD74 through shRNA , the knockdown of CD74 itself was sufficient to inhibit CD4+ T cell recognition ( Figure 8 ) . In addition to serving as a chaperone for MHC-II , CD74 has been reported to play an essential role in B cell maturation [33] , which involves activation of transcription mediated by p65 member of the NF-κB family [34] . These two functions of CD74 are genetically separable and map to different regions of the protein [35] . More recently CD74 has been identified as a receptor for MIF , and to promote cell survival and proliferation [26] . Binding of MIF to CD74 triggers activation of the p65 member of the NF-κB family , which in turn trans-activates Bcl-2 family genes , thereby providing the cells with increased survival capacity [27] . Furthermore , antibodies that block MIF/CD74 interaction cause growth inhibition and induction of apoptosis in B-cell lines [36] . BZLF1 is known to inhibit NF-κB p65 activity [37] and is toxic for all CD74+ cell lines used in our experiments , a phenomenon that correlated with downregulation of the Bcl-2 and Bcl-xl anti-apoptotic proteins ( Figure 3C , 4C ) . It is notable that we observed no toxicity of BZLF1 in the epithelial cell line , HEK-293 , which lacks expression of MHC-II and CD74 . The kinetics of the toxicity of BZLF1 in the EBV-negative Akata-A3 B cell line is such that BZLF1-transfected cells survive only 2 days ( Figure 3A ) . This contrasts with what is observed during the normal physiological process of lytic cycle in the EBV-positive parental Akata line , where cell viability is maintained for at least 4 days after expression of BZLF1 [29] . EBV encodes two vBcl-2 homologs , both of which are expressed early following initiation of lytic cycle by EBV . The best characterized vBcl-2 is BHRF1 , a potent anti-apoptotic protein that clearly enhances survival of B lymphocytes [21] and whose molecular mechanisms are beginning to be elucidated [38] . In contrast , it is unclear whether the second vBcl-2 , BALF1 , actually functions to modulate apoptosis [39] . In the present study , we showed that BHRF1 alone is able to moderate BZLF1 toxicity to an extent that is consistent with the enhanced survival period of cells entering lytic cycle . In this study we have shown for the first time that the MHC-II antigen presentation is impaired during lytic infection of normal B cells ( Figure 1 ) . About 80 antigens are expressed in the EBV lytic cycle , representing a large pool of potential target antigens as reflected in the broad repertoire of EBV-specific CD4+ T responses identified , including some clones with cytotoxic activity to these EBV antigens [9] , [10] . Therefore , impairment of MHC-II antigen presentation is likely to be crucial for the lytic cycle cells to survive long enough to generate infectious virus progeny . In addition to T cell responses to newly-synthesized early and late antigens in lytic cycle , there are also responses to pre-existing MHC-II/peptide complexes from latent antigens expressed at the time of initiation of lytic cycle . In this context , it is interesting to note that recognition of EBNA1 , which is expressed during both latent and lytic infection , by specific CD4+ T cells is also impaired following expression of BZLF1 and induction of lytic cycle ( Figure 1D ) . As BZLF1 is the first EBV antigen to be expressed during lytic cycle , a process that can be sustained for several days before cell death occurs [29] , its immune-evasion functions may be paramount in EBV's strategy for attenuating anti-viral responses . In this context , the impairment of the MHC-II antigen presentation pathway by BZLF1 adds to other previously reported immune-modulating properties of BZLF1 , notably; inhibition the IFN-gamma signaling pathway [40] and TNF-alpha activation [41] , [42] by down-regulation of IFN-γ receptor and TNF-R1 . However , with regards to modulation of MHC-II antigen presentation it is likely that multiple EBV genes will cooperate to evade immune CD4+ T cell responses , as is seen with MHC-I antigen presentation to CD8+ T cells [43] . The exonuclease/host shut-off protein , BGLF5 , expressed in early lytic cycle may contribute by degrading MHC-II mRNA transcripts [2] , [3] and the late BZLF2 glycoprotein , gp42 , may contribute by binding to MHC-II molecules and sterically inhibiting recognition by the T cell receptor of immune CD4+ T cells [16] . In summary , this work provides a new paradigm for viral immune evasion of MHC-II presented antigen . Targeting CD74 expression is sufficient to substantially impair MHC-II presentation of antigenic peptides even when levels of MHC-II DR molecules are barely affected . However , as CD74 also serves as an important regulator of cell survival , fresh insight is provided as to the role of vBcl-2 during lytic cycle in B cells . It is widely accepted that BHRF1 prolongs cell survival during lytic cycle to allow sufficient time for production and accumulation of new infectious virions . Now , we suggest that BHRF1 also plays a pivotal role in enabling BZLF1 to attenuate recognition by CD4+ T cell responses .
A derivative of the DOX-dependent expression vector pRTS-1 [44] was kindly provided by Dr J Mautner , Munich; BZLF1 and a reverse BZLF1 sequence as control were introduced into the vector by standard DNA cloning procedures to create vectors pRTS-CD2-BZLF1 and pRTS-CD2-control . The EBV lytic genes BZLF1 , BZLF2 , BALF4 were also subcloned into the EcoRI/NotI sites of pCDNA3-IRES-nls-GFP vector . All plasmids were verified by restriction digest and sequence analysis . The pCDNA3-cyto-EBNA1 plasmid was described previously [20] . Transient transfection of MJS cells with plasmid DNA was routinely performed using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions . Targets for the T cell recognition assay clone were generated by co-transfection of HLA-DRB5*01 MJS cells with a cyto-EBNA1 expression plasmid and IRES-GFP , BZLF2-GFP , BALF4-GFP or BZLF1-GFP expression plasmids . LCLs were established using the reference B95 . 8-based recombinant lacking the BZLF1 gene ( BZLF1KO ) [45] . A doxycycline ( DOX ) -inducible BZLF1 expression vector , pRTS-CD2-BZLF1 , or control vector with the reverse BZLF1 sequence ( pRTS-CD2-control ) were introduced into LCLs or Akata-A3 by electroporation of 10 µg plasmid DNA into 107 cells in OptiMem medium ( Invitrogen ) at 280 V and 960 µF using a Biorad electroporation apparatus . Transfected cells were cultured in RPMI-1640 supplemented with 10% fetal calf serum ( FCS ) . After 24 h , the transfected cell population was enriched by staining with OX34 antibody to rat CD2 , and positively selected by magnetic cell sorting with anti-mouse IgG2a/b Microbeads and LS columns ( Miltenyi Biotech ) according to the manufacturer's guidelines . Cells were thereafter expanded and maintained in RPMI 1640 medium supplemented with 10% FCS . BZLF1 expression was induced by addition of 200 ng/ml DOX for 24 h , and the induced cells were positively selected by magnetic cell sorting with anti-NRFR Microbeads and LS columns ( Miltenyi Biotech ) . The purity of the sorted cells was checked with Beckman Coulter XL flow cytometer . The MJS ( Mel JuSol ) melanoma-derived cell line [46] was maintained in RPMI 1640 medium ( Gibco BRL ) supplemented with 10% FCS . HLA-DRB5*01 expressing MJS cells were generated by transduction with a DRB5*01 retrovirus vector; a HLA-DRB5*01 β chain gene was cloned into retroviral expression plasmid pQCXIN ( Clontech ) by standard methods . Vesicular stomatitis virus-pseudotyped retrovirus particles were produced in GP2-293 cells co-transfected with the pVSV-G envelope vector . Virus in the culture supernatant at 72 h was concentrated by ultracentrifugation and used to infect 5×105 target cells overnight . Infected cells were selected with G418 ( Invitrogen ) . BHRF1 retroviral constructs were engineered by cloning the cDNA encoding BHRF1 into the pLZRS retroviral vector . Immediately downstream from the inserted BHRF1 gene lies an IRES sequence and the marker gene , a truncated nerve growth factor ( ΔNGFR ) . Vesicular stomatitis virus-pseudotyped retrovirus particles were produced as above and used to transduce MJS cells . Transduced cells were magnetically sorted using MACS NGFR-specific beads as directed by the manufacturer ( Miltenyi Biotech ) . The lentivirus plasmids containing a sequence of CD74-specific shRNA or a sequence of scrambled shRNA were purchased from Santa Cruz Biotechnology . Vesicular stomatitis virus-pseudotyped lentivirus particles were produced in FT-293 cells co-transfected with the pVSV-G and Gag-Pol expressing vectors . Virus in the culture supernatant at 72 h was concentrated by ultracentrifugation and used to infect 5×105 target cells overnight . Infected cells were selected with puromycin ( Sigma ) . Cells were starved by culturing 107 in 15 ml methionine-free RPMI medium supplemented with 10% dialysed FCS for 1 h at 37°C , then labeled for 15 min with 200 µCi of 35S protein labeling mix ( PerkinElmer ) in a final volume of 1 ml . After two washes with chase medium ( normal RPMI medium supplemented with 10% FCS ) , the cells were resuspended at 2×106 cells/ml and chased at 37°C for the times indicated . Samples containing 2×106 cells were lysed in 400 µl of NP-40 buffer ( 0 . 5% Nonidet P-40 , 5 mM MgCl2 and 50 mM Tris-HCl , pH 7 . 5 ) with protease inhibitor cocktail ( Sigma ) at 4°C for 45 min . Nuclei and insoluble debris were removed by centrifugation , and the supernatants were precleared; first with 1 . 2 µl normal mouse serum and 20 µl Dynabeads Protein A ( Invitrogen ) for 2 h at 4°C , and then with 20 µl Dynabeads Protein A and 20 µl Dynabeads protein G at 4°C overnight . The precleared lysates were immunoprecipitated for 2 h with 1 µg of mouse anti-CD74 and 20 µl Dynabeads Protein A plus 20 µl Dynabeads protein G , before washing the beads four times with NET buffer ( 0 . 5% NP-40 , 150 mM NaCl2 , 5 mM EDTA and 50 mM Tris-HCl , pH 7 . 5 ) and eluting by boiling in reducing gel sample buffer for 5 min . Finally , the samples were separated by SDS-PAGE on 12% Bis-Tris NuPage mini-gels with MOPS buffer ( Invitrogen ) . After the gels were fixed and dried , they were exposed to autoradiographic film . Goat antibodies to calregulin ( sc6467 ) , mouse anti-Bcl-xl , the mouse anti-CD74 and the mouse anti-DRα were purchased from Santa Cruz Biotechnology . Clone 124 mouse antibody to Bcl-2 [47] was a kind gift from the late David Mason . Antibodies to EBV antigens are described elsewhere [48] . For flow cytometry experiments , phycoerythrin ( RPE ) -conjugated goat anti-mouse IgG antibodies were purchased from AbD Serotec . Cell surface expression of MHC-II DR or CD74 on viable cells was determined by staining with PE-conjugated anti-DR ( AbD Serotec ) or anti-CD74 primary antibodies followed with RPE-conjugated goat anti-mouse IgG2a ( AbD Serotec ) . Intracellular staining to detect cells expressing nuclear BZLF1 was performed on 106 cells that were fixed using 100 µl of Ebiosciences Intracellular ( IC ) Fixative for 1 h on ice , followed by permeabilisation through the addition of Triton X-100 to a final concentration 0 . 2% for a further 30 min incubation on ice . After extensive washing with PBS , the cells were stained with 1 µg/ml of either MAb BZ . 1 ( anti-BZLF1 ) or with an IgG1 isotype control MAb for 1 hr at 37°C , followed by a 1∶50-dilution of RPE-conjugated goat anti-mouse IgG1 antibody ( AbD Serotec ) for 1 hr at 37°C . Stained cells were analyzed on Beckman Coulter XL flow cytometer , and the data processed using Flowjo software ( Tree Star ) . Total cell lysates were denatured in reducing sample buffer ( final concentration: 2% SDS , 72 . 5 mM Tris-HCl pH 6 . 8 , 10% glycerol , 0 . 2 M sodium 2-mercaptoethane-sulfonate , 0 . 002% bromophenol blue ) , then sonicated and heated to 100°C for 5 min . Solubilized proteins equivalent to 105 cells/20 µl sample were separated by sodium dodecylsulfate polyacrylamide gel electrophoresis ( SDS-PAGE ) on 4–12% acrylamide gradient Bis-Tris NuPage mini-gels with MOPS running buffer ( Invitrogen ) . Following electroblotting to polyvinylidene difluoride membranes , immunoblotting with specific primary antibodies followed by detection with appropriate alkaline phosphatase-conjugated secondary antibodies and a CDP-Star™ chemiluminescence detection kit ( Tropix , Applied Biosystems ) was performed as previously described [5] . T cells were grown in 10% FCS in RPMI-1640 medium supplemented with 30% supernatant from the IL-2-producing MLA 144 cell line and 50 U/ml recombinant IL-2 . The effector CD4+ T cell clones ‘LEK’ ( specific for amino acids 126–140 of BXLF2 and restricted through DRB5*01 ) and ‘SNP’ ( specific for amino acids 474–493 of EBNA1 and restricted through DRB5*01 ) were generated as described elsewhere [10] , [20] . The effector CD8+ T cell clone ‘HPV’ ( specific for aa 407–417 of EBNA1 and restricted through MHC-I B35 . 01 ) was described elsewhere [49] . The capacity of CD4+ and CD8+ T cell clones to recognize target LCLs or MJS cells was measured by IFNγ ELISA ( Endogen ) . Briefly , 104 effector T cells were incubated for 18 h at 37°C in V-bottom microtest plate wells with 105 target cells , before assaying the supernatants for IFN-γ release by ELISA ( Endogen ) in accordance with the manufacturer's recommended protocol . Total RNA was isolated from cultured cell lines using QIAGEN RNeasy kit and treated with DNase I ( Turbo DNA-free kit; Ambion ) . Quantitative reverse-transcription polymerase chain reaction ( QRT-PCR ) assays for DRA , CD74 , Bcl-2 and BCl-xl were performed with TaqMan® Gene Expression Assays ( applied biosystem ) , duplexed with GAPDH assays for normalization .
|
Epstein-Barr virus ( EBV ) is a herpesvirus and an important human pathogen that can cause diseases ranging from non-malignant proliferative disease to fully malignant cancers of lymphocytes and epithelial cells . The persistence of EBV in healthy individuals relies on the balance between host immune responses and viral immune evasion . As CD4+ immune T cell responses include both helper and cytotoxic functions , viral mechanisms for interfering with MHC class II antigen presentation to CD4+ T cells have the potential to greatly influence the outcome of viral infections . Our work on Epstein-Barr virus provides a new paradigm for viral immune evasion of MHC-II presented antigen by targeting CD74 . CD74 is a dual function protein; it serves as a surviving receptor as well as a chaperone for MHC-II antigen presentation . Therefore , downregulation of CD74 as a T cell evasion strategy comes at the cost of potentially inducing cell death . However , EBV also encodes a vBcl-2 to attenuate the toxicity associated with reduced CD74 , thus enabling the immune-impairment function to be effected . We expect that future studies will identify other viruses utilizing a similar strategy to evade CD4+ immune T cell responses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"death",
"viral",
"classification",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"dna",
"viruses",
"major",
"histocompatibility",
"complex",
"viral",
"immune",
"evasion",
"biology",
"antigen",
"processing",
"and",
"recognition",
"cell",
"biology",
"virology",
"molecular",
"cell",
"biology"
] |
2011
|
Epstein-Barr Virus Evades CD4+ T Cell Responses in Lytic Cycle through BZLF1-mediated Downregulation of CD74 and the Cooperation of vBcl-2
|
Whole-brain recordings give us a global perspective of the brain in action . In this study , we describe a method using light field microscopy to record near-whole brain calcium and voltage activity at high speed in behaving adult flies . We first obtained global activity maps for various stimuli and behaviors . Notably , we found that brain activity increased on a global scale when the fly walked but not when it groomed . This global increase with walking was particularly strong in dopamine neurons . Second , we extracted maps of spatially distinct sources of activity as well as their time series using principal component analysis and independent component analysis . The characteristic shapes in the maps matched the anatomy of subneuropil regions and , in some cases , a specific neuron type . Brain structures that responded to light and odor were consistent with previous reports , confirming the new technique’s validity . We also observed previously uncharacterized behavior-related activity as well as patterns of spontaneous voltage activity .
Measuring activity simultaneously in the whole brain is critical to understanding how different brain regions interact to process and control sensory inputs , internal states , and behavior . Whole-brain recordings not only reveal which regions are involved in which functions and with what network dynamics but also help to interpret the effects of a targeted intervention ( e . g . , a lesion or a local alteration with optogenetics ) on the whole network and give context to local electrophysiology recordings . Furthermore , they are necessary to characterize global changes affecting the brain on a large scale ( such as different behavioral states ) and detect patterns of activity involving distant regions . This is already possible in humans , for which functional magnetic resonance imaging ( fMRI ) has opened a new chapter in the study of brain activity underlying behavior , but this technique has low spatial and temporal resolution . In animals , techniques for imaging a whole brain so far have allowed us to record activity at much higher resolutions but are still orders-of-magnitude slower than neuronal electrical activity . For example , recent reports of scanning-based whole-brain imaging in zebrafish and Drosophila larvae had a frame rate of 12 Hz [1] and 5 Hz [2] , respectively . By contrast , light field microscopy [3–9] makes it possible to image large volumes of scattering brain tissue at more than 100 Hz . In this study , we leverage this technique to record large-scale activity in the brain of behaving adult fruit flies . We present a method to optically access the fly’s brain while enabling it to retain the ability to walk or groom . We show that the near-whole brain can be imaged with a 20x objective at a frame rate up to 200 Hz and fluorescence recorded from pan-neuronally expressed calcium ( GCaMP6 [10] ) or voltage ( ArcLight [11] ) probes . We present rich data sets of near-whole brain activity and behavior as well as two analysis methods . First , we map activity for specific stimuli and behaviors with short time-scales; for example , we compared activity when the fly rested , walked , and groomed . Second , we apply a computational method ( principal component analysis [PCA] , followed by independent component analysis [ICA] ) to extract components representing spatially distinct sources of activity [6 , 12 , 13] . We show that these sources correspond to subneuropil areas or processes from small populations of neurons that are anatomically well characterized , and we compare their responses to flashes of light or odor puffs with those in literature reports of experiments done on restricted regions . Additionally , by using this method , we discovered neuronal projections whose activity correlated with turning as well as previously unreported patterns of spontaneous voltage activity .
We first fixed a fly’s head by rotating it 45 degrees or more around the transversal axis to decrease the depth of the volume imaged and to improve access to the brain . We then exposed the brain while keeping the eyes , antennae , and legs intact and clean ( see Methods section and S1 Fig ) . A ball was placed under the fly’s tarsi so that it could typically rest , walk , and groom . We imaged the fly brain’s fluorescence using light field microscopy . As shown in Fig 1A , we modified an upright epifluorescence microscope ( equipped with a 20x 1 . 0 numerical aperture [NA] or a 40x 0 . 8 NA objective ) by adding a microlens array at the image plane of the objective and placing the camera sensor at the image plane of the microlens array through relay lenses . We recorded light field images continuously with a high-speed scientific complementary metal-oxide-semiconductor ( sCMOS ) camera up to 100 Hz for GCaMP6 and 200 Hz for ArcLight ( using the middle of the camera sensor ) . We then reconstructed the volumes—typically 600 x 300 x 200 μm3 to encompass the whole brain ( Fig 1B ) —using the volumetric deconvolution method for light field microscopy described in [3] . Note that unlike other microscopy techniques that are based on scanning ( e . g . , two-photon , confocal , or light-sheet microscopy ) , excitation light illuminates the entire brain continuously , and all the photons emitted in the numerical aperture of the objective are used to reconstruct the image ( minus an approximately 40% loss through the objective , tube lens , microlens array , and relay lenses ) . This maximizes the number of photons collected ( and thus information about brain activity ) per units of volume and time . We used 2-μm fluorescent beads embedded in a gel to measure the point spread function and found that it widens with distance from the focal plane , varying from 3 . 5 to 12 μm laterally and from 6 to 35 μm axially for 20x 1 . 0 NA objective and varying from 2 to 7 μm laterally and from 4 to 22 μm axially for 40x 0 . 8 NA objective ( S2 Fig and theoretical expression in [3] ) . As shown in S3 Fig ( presenting ArcLight’s baseline fluorescence ) and below , this resolution was sufficient to recognize neuropil structures and extract activity from subneuropil compartments . S1 and S2 Movies present maximum z projections of near-whole brain activity ( after preprocessing as described in S4 Fig and the Methods section ) when stimuli were presented to the fly . Fig 2C and S5C Fig show maps of the response to stimuli . We found strong increases in activity at the onset of puffs of odor and flashes of UV light in specific parts of the brain ( see S5–S7 and S9 Figs ) , in accordance with previous reports in the literature: the strongest responses to light involved the optic lobes , optical glomeruli , and posterior slope , and the responses to odor involved the antennal lobe and most of the dorsal neuropils including the lateral horn , superior neuropils , and the mushroom body ( S5C , S6 , S7 and S9 Figs ) . The global map of the response to stimuli was similar for calcium ( GCaMP6 ) and voltage ( ArcLight ) activity ( S5C and S9 Figs ) . We also examined near-whole brain activity in the absence of external stimuli ( i . e . , spontaneous behavior ) , which consisted of walking , grooming , and resting ( see S3–S7 Movies ) . Most strikingly , the brain was more active on a global scale when the fly walked than when it rested or groomed ( see S3 Movie , Fig 2A , and S5A , S8 , and S9 Figs ) , and global activity ( average ΔF/F ) was correlated with walking ( R2 = 0 . 37 +/− 0 . 19 , average +/− STD , N = 6 ) . To verify that this response was linked to walking rather than the optic flow from the ball , we repeated the experiment with a visually impaired norpA mutant fly and again found a global increase during walking in comparison with rest ( see S4 Movie and S5A Fig ) . In contrast , we found only local activation in the region of the saddle , wedge , and antennal mechanosensory and motor center ( in four out of five flies ) during grooming . To investigate whether the global increase during walking was coming exclusively from neurons expressing one type of neurotransmitter or neuromodulator , we performed the same experiments with more restricted lines . We also found a global increase when GCaMP6f was expressed in cholinergic neurons , which form the majority of excitatory neurons in the fly brain ( R2 = 0 . 49 for the correlation of average ΔF/F with walk ) ( S5 Movie and S5A Fig ) . When GCaMP6f was expressed in dopamine neurons only ( with TH-Gal4: S6 Movie , Fig 2A , and S5A Fig ) , we observed a strong large-scale increase of activity tightly locked with walking . We also found surprisingly little activity during resting or grooming , apart from the mushroom body area . This increase was observed in flies expressing GCaMP6f in both dopamine and serotonin neurons ( TH-Gal4 and DDC-Gal4 ) , but in that case , we also observed more background activity unrelated to behavior ( S7 Movie and S5A Fig ) . We investigated the difference in activity when the fly turned left compared to right ( Fig 2B and S5B Fig ) . Although there was a strong variability from fly to fly that will need to be characterized in future studies , we observed antisymmetric patterns in the ventral areas and in the lateral superior protocerebrum ( as indicated by the arrows ) in all flies . We used a combination of statistical methods to extract maps and time series of spatially distinct sources of activity ( see Methods section for details ) . Briefly , we first applied PCA ( using singular value decomposition ) to find maps of correlated activity and to reduce dimensionality . We then applied ICA to unmix the PCA maps and find sparse functional regions ( see analysis pipeline in S4 Fig ) . Fig 3 shows z-stacks containing all components ( z-score > 3 ) from pan-neuronal GCaMP6 recordings ( see also additional flies in S10 and S11 Figs ) . We also performed the same analysis with flies that pan-neuronally expressed activity-independent green fluorescent protein ( GFP ) to help identify artifact-related components—movement , aliasing , or noise , as shown in S12 Fig . Even though PCA and ICA are mathematical algorithms that make minimal assumptions about the brain , most functional maps matched well with anatomical structures . We aligned the brain with an anatomical template [14] using landmarks registration and automatically sorted the components by brain region ( Fig 4A ) . In Fig 4B , the left column presents the component’s thresholded maps , whereas the right column presents central complex structures or neuronal processes from [15] , assembled using Virtual Fly Brain [16] . Several subneuropil regions are recognizable from the shape of the maps ( e . g . , protocerebral bridge glomeruli , ellipsoid body rings , and fan-shaped body layers ) . For some components , the combination of subneuropil regions present in the map allowed to assign it to a specific major neuron type ( see Fig 4B , bottom part ) . For example , z-scored maps with signals spanning both the alpha and the beta mushroom body lobes likely resulted from activity in alpha/beta Kenyon cell axons , maps with signals spanning both the alpha’ and the beta’ mushroom body lobes likely resulted from activity in alpha’/beta’ Kenyon cell axons , whereas maps with signal in the gamma lobe likely resulted from activity in the gamma Kenyon cell axons . Likewise , z-scored maps containing signals in one antennal lobe subneuropil region ( with a glomerulus-like shape ) , in the calyx and in the lateral horn likely resulted from the activity of antennal lobe projection neurons . Maps spanning a medulla column and a lobular layer were likely from transmedullar neurons . Finally , the maps of the components matching the protocerebral bridge glomeruli also often contained radial parts of the ellipsoid body ( e . g . , Fig 4B , left column , top panel ) , suggesting that these components might originate from tile or wedge neurons [18 , 19] . The component’s time series ( resulting from PCA/ICA or from region of interest [ROI] ΔF/F averages; S13 and S15–S17 Figs ) were consistent with previous reports of activity from the brain structures identified in the components’ maps . Most of the components responding to the onset and/or offset of light were in the optic lobe [20] ( S14 and S15 Figs ) . In contrast , components responding to puffs of odors were mostly in the antennal lobe , the lateral horn , and the mushroom body [21] ( S14 and S16 Figs ) . Components likely representing the activity of antennal projection neurons were spontaneously active in the absence of odor , but some of them became more active when the odor was presented ( S14C and S16A Figs ) , consistent with previous reports [22] . In addition , regions in the lateral protocerebrum , the posterior slope , the antennal mechanosensory and motor center ( AMMC ) , the saddle , and the protocerebral bridge were strongly active when the fly walked ( S17 Fig ) . This is consistent with previous anatomical studies; the projection from the descending neurons are most dense in the posterior slope , ventrolateral protocerebrum , and the AMMC [23] . Some of these walking-related components were strongly correlated with turning left or right ( see Fig 5; Pearson’s R = 0 . 63 ± 0 . 16 , mean ± SD , N = 12 cyan or yellow components , and Pearson’s R = 0 . 59 ± 0 . 20 , mean ± SD , N = 8 blue or red components ) . Their distribution in the brain as well as strong structural characteristics ( e . g . , tracks forming an inverted “V” shape , fine dorsal claw-like neurites ) suggest that they were generated by the same neurons across flies . Furthermore , the V-shaped components always preceded the other component correlated with turning toward the same direction—the peak of the cross-correlation function was at 68 ms for GCaMP6F ( SEM = 11 ms , N = 6 pairs of components ) . As shown in S18 Fig , we found similar components when using a Cha-Gal4 driver instead of a pan-neuronal driver ( average cross-correlation peak time was 80 ms ) , which suggests that those neurons are cholinergic . Note that these components are mostly present in the posterior slope , as are neurons involved in turning during flight [24] . Fig 6 and S19 Fig show components obtained when using a more restricted driver , expressing in dopamine neurons only: TH-Gal4 ( data are the same as in S6 Movie ) . As shown in Fig 6A , in agreement with our observation from S6 Movie , most components were tightly correlated with the fly walking ( forest green traces interleaved with the components’ traces ) . Fig 6B reproduces some of the maps in Fig 6A along with anatomical maps of single dopaminergic neurons from the Virtual Fly Brain database [16] . Some maps had unequivocal anatomical counterparts . For example , the first three maps matched well with the anatomy of processes from dopaminergic PPL1 neurons innervating mushroom body vertical lobe compartments , each component thus corresponding to only one or two cells per hemisphere [25] . As Fig 7 and S20 Fig demonstrate , voltage recordings with ArcLight also gave rise to maps portraying specific neuropils ( and clearly distinguishable from artifacts shown in S21 Fig ) . As S22 Fig shows , the number of components per brain region was typically smaller than it was for GCaMP6—we extracted an average 174 ( STD = 68 , N = 12 ) activity-related components ( i . e . , not noise or movement artifacts as detailed in the Methods section ) from GCaMP6 recordings and 54 ( STD = 14 , N = 6 ) from ArcLight recordings , probably because of the probe’s lower signal-to-noise ratio . However , ArcLight components were similar to those found with GCaMP6: in the optic lobe , some components responded to the onset and/or offset of light , with various degrees of habituation . In the posterior slope , we found peaks at the onsets of light . We also recorded large peaks of activity in the ventrolateral protocerebrum . Finally , we found components in the antennal lobe , lateral horn , and mushroom body responding to odor ( S23 Fig ) . One clear difference between voltage and calcium was the presence in the ArcLight data of slow , spontaneous switches between the up and down levels of activity for components in a nodulus and contralateral part of the protocerebral bridge ( Fig 8 ) . We did not observe those components in controls in which GFP was expressed pan-neuronally . Furthermore , although some movement artifacts could generate slow fluctuations in those regions , we only observed asynchronous steps when using ArcLight ( see difference between the two sides in S24 Fig ) . Other patterns of spontaneous activity included oscillations in the antennal lobe and lateral horn and fast , ongoing activity in the ellipsoid body and the protocerebral bridge ( see S25 Fig for an experiment in which different time scales of spontaneous activity were detected ) . More work is necessary to establish the conditions and consistency of these patterns of activity . Note that time series from single trials had high enough signal-to-noise ratios to detect graded potentials ( e . g . , components in the optic lobe in response to the onset and/or offset of flashes of light ) and spike-like signals , possibly containing several action potentials ( e . g . , spontaneous activity and odor response for components in the antennal lobe , mushroom body , and lateral horn [Fig 7 and S25 Fig] ) , which are consistent with previous literature [21] [20] . Spike-like signals were particularly clear in a more restricted driver for dopamine and serotonin neurons ( i . e . , TH-Gal4 and Dopa-decarboxylase or DDC-Gal4 ) , as shown in S26 Fig . Recent studies have shown that large-scale brain imaging in flies is possible with other imaging methods . Mann and colleagues [26] used a high-speed , two-photon microscope to image the brain with higher resolution but at a slower rate ( 1 Hz ) in the absence of stimuli or behavior . We applied our analysis pipeline to these data to compare it to the results from the light field microscope ( see S27 Fig ) . Out of a total of 23 activity-related components ( compared to an average of 174 for our light field recordings , as described above ) , eight could be interpreted as cell bodies that could not be extracted at this resolution from the light field data . Other components—components covering the antennal lobe , mushroom body , and lateral horn; components in the pars intercerebralis; and components in the antennal lobe glomerulus , calyx , and lateral horn ( likely representing activity from antennal lobe projections neurons ) —had similar spatial distribution as components extracted from the light field data . Faster techniques have also been used to image large-scale activity in flies: [27] used a Bessel beam to image 25% of the brain at 3 . 6 Hz , and [28] used a light sheet approach ( swept confocally aligned planar excitation [SCAPE] ) to image a large portion of the brain at 10 Hz . To test whether the fast-frame rate of the light field method enabled detecting signals otherwise undetectable with slower imaging methods , we subtracted the data smoothed over 100 ms , thus revealing only the activity above 10 Hz . All GCaMP6f data sets ( N = 7 ) maintained some activity-related components ( e . g . , S28 Fig ) that could be different in different flies . When we did the same analysis with ArcLight data , four out of seven flies had at least one activity-related component . Note that more information from fast activity could be present in the data , but the low signal-to-noise ratio makes it difficult to detect it with PCA/ICA .
The method permits imaging of the near-whole brain with a frame rate of 200 Hz; however , the time response of the probes we used in this study is slower ( rise time to spike peak of approximately 0 . 15 seconds for GCaMP6f and approximately 0 . 1 seconds for ArcLight in our hands ) . Fast activity can still be detected , but the probe’s response imposes a temporal filter on the underlying activity . The ability to record high signal-to-noise transients with such a high frame rate suggests that the light field microscope will be suited to measure activity from faster probes . This will help bridge the gap between the current fast local methods using microelectrodes ( e . g . , recording spikes and fast oscillations ) and slower large-scale methods ( e . g . , calcium imaging ) . The excitation light excites the eye photoreceptors , thus affecting the fly’s ability to see as well as potentially changing brain activity states and behavior . The fly’s blue light receptors could be genetically removed and potentially replaced by another receptor , such as one for UV light , if a future researcher wanted to study brain responses to visual inputs without the strong background activation from the excitation light . For applications that do not necessitate visual inputs , blind fly mutants ( e . g . , a norpA , cryptochrome mutant ) could be used to affect all light detection in the brain . Finally , blue-light excitation probes could be replaced with probes with longer excitation wavelengths . Dissection could have affected the fly’s state . Removing the cuticle on the back of the brain could affect brain activity by activating nociceptor neurons ( e . g . , those in the bristles ) . Dissection could also have affected the fly’s global health state . Indeed , we found that flies expressing ArcLight pan-neuronally were usually less active after dissection than they were before , making it difficult to obtain reliable behaviors with this genotype . Finding the optimal recovery time after dissection could help minimize these effects . Imaging the brain of nondissected flies genetically modified to have a cuticle with low absorbance ( e . g . , yellow flies ) could also help characterize the effects of the dissection . Although the fly could still move its legs , abdomen , and ( to a limited extent ) its wings , the immobilization of its head , proboscis , and thorax could have affected brain activity and behavior by imposing unnatural constraints . Furthermore , the fly’s head was tilted more than 45 degrees in comparison to its natural position in order to better align the thinner part of the brain to the z-axis . This helped minimize the loss of resolution with depth . We observed a seemingly natural behavior in this configuration ( with alternations between grooming and walking as free flies do ) ; however , we sometimes found the fly displaying unnatural behaviors , such as pushing the ball away ( see behavior and corresponding brain activity in S5 Movie ) or touching the holder with its legs . Another problem resulting from immobilizing the fly’s head was the lack of coupling between the stimuli position and the fly’s movement that would normally occur in a natural setting . This problem can be solved by using a virtual reality setup in a closed loop configuration ( e . g . , using the movements of the ball to change the stimuli position ) . The whole procedure made it impractical to obtain data from a large number of flies . Even with practice , fly preparation remained challenging to the extent that it was difficult to obtain more than one good preparation per day . Another factor limiting data production—but that should be less and less of a problem as computing cost continues to drop—was the reconstruction step , which takes approximately 10 hours on a cluster of 16 graphical processing units ( GPUs ) for a data set of 60 GB ( which corresponds to approximately 1 minute of recording at 200 Hz ) . This method is thus suited for studying complex spatiotemporal patterns and identifying neurons and brain structures in a few trials and flies but not for larger studies , such as genetic screens . Detecting sources from the raw light field data could help reduce the cost of reconstruction . For example , anatomical maps could be transformed back to a light field image and used as seeds for the source extraction algorithm [8] . Although we can observe the whole central brain ( though the access to the gnathal ganglia depended on the quality of the preparation ) as well as a large part of the optic lobes ( typically the lobula and most of the medulla ) , we cannot observe all of the fly neurons . In particular , the ventral cord in the fly thorax is not accessible with the current setup . Imaging the ventral nerve cord in addition to the brain might be feasible with an objective with a larger field of view and the appropriate dissection preparation [38] . As the brain contains approximately 105 neurons , and we record , at most , several hundred activity-related components , we are far from obtaining recordings from all neurons . This could be due to various reasons . First , some neurons might be silent and thus undetectable by our algorithm , which is based on temporal changes . Second , the signal-to-noise ratio might be insufficient for PCA and ICA to detect the activity in some processes . To increase the signal-to-noise ratio and obtain more components , future researchers could use a more sensitive probe , record longer time series , or use faster probes to obtain more temporal information . Third , several neurons might contribute to one component . Indeed , neurites with similar presynaptic inputs and thus similar activity patterns will likely have similar geometry , making them indistinguishable to the algorithm . Additionally , the resolution of the microscope is , in general , larger than individual somata and neurites . In particular , the low-axial resolution far from the focal plane ( which could be improved with improved light field techniques [9] ) makes it difficult to sort out the activity from regions that are close to each other , such as the antennal lobe and the lateral accessory lobe or the protocerebral bridge and the antler . However close to the focal plane , the functional maps were the same scale as functional maps obtained with higher-resolution microscopy techniques ( e . g . , in the fan-shaped body [17] and the lateral horn [39] ) or regions known to be functional units ( e . g . , antennal lobe glomeruli and ellipsoid body wedges and tiles ) . Using a second color and complementary drivers ( e . g . , drivers for excitatory versus inhibitory neurons or drivers for main neurotransmitters versus drivers for neuromodulatory neurons ) could increase the number of components that can be detected . The identification of anatomical structures could also be improved . Currently , the registration of the light field data with the anatomy is done using landmark registration . This method is imprecise in brain areas that lack clear landmarks , such as the ventral areas . Concurrently imaging the brains using a different microscopy technique with higher resolution could help detect more landmarks or make it possible to use different registration techniques . Another way to improve registration would be to use driver lines for specific regions expressing a fluorophore with another color ( e . g . , red fluorescent protein [RFP] ) . Expressing activity probes with drivers for specific regions could also be used to verify the presence of components in those regions . Automating the search for matches between components’ maps and neurons in large databases such as Flycircuit or Virtual Fly Brain would help to get to the level of neurons rather than brain regions . The maps obtained using PCA and ICA can have regions with both positive and negative values , but this study has ignored the negative parts of the maps . More work is necessary to characterize the meaning of those negative values . In particular , neurons underlying the positive part of the maps could be inhibiting the neurons underlying the negative part of the maps or vice versa . The PCA/ICA algorithm used here helps to unmix neural activity from movement artifacts or from other overlapping processes as well as scattered activity coming from other parts of the brain ( see S12 , S13 , S15–S17 , and S21 Figs ) . However , the interpretation of these time series is not straightforward , as there is no guarantee that the algorithm will extract the full neural activity from one source . Furthermore , the imperfect spatial separation of the sources can lead to artifacts in the case of strong synchronous fluorescence changes in large parts of the brain . For example , a negative signal can be present for components in the optic lobe after the odor is presented . As this signal is not present when measuring fluorescence in the region of interest delimited by the z-scored maps ( putting all pixel values below three times the standard deviation to zero ) or when applying PCA and ICA in the region of the optic lobes only , it is likely due to an imperfect separation of the optic lobe components from the regions in the middle of the brain where fluorescence strongly changes in response to the odor . Indeed , the maps for the optic lobe components have small negative values in the mushroom body and antennal lobe areas . To recognize these artifacts , observing both the unmixed time series and the fluorescence of ROI is thus advisable , as done in S13 and S15–S17 Figs . Using different algorithms might help prevent these artifacts . For example , non-negative matrix factorization could better separate overlapping processes and avoid spurious negative values in the maps; however , in our hands , the components were less localized and thus more difficult to interpret than with PCA/ICA . Movement correction with 3dvolreg can be imperfect and even can , in some cases , introduce additional artifacts when strong fluorescence changes are present in large parts of the brain . Furthermore , the algorithm uses rigid registration and does not correct for local deformations . Although we partly subtract these artifacts at the PCA and ICA stages of the analysis , they can complicate the interpretation of some of the time series . Better movement-correction methods with a limited sensitivity to fluorescence changes ( such as sparse and low-rank decomposition [7 , 40] ) and nonrigid registration [41–43] as well as using an activity-independent fluorophore in another color channel as a reference would improve the reliability of the time series . Despite these limitations , the methods presented in this study can be used as a functional screen to identify brain regions and neurons involved in processing any stimulus or behavior that a fly can perform under the microscope . Furthermore , complementary to screens using activation or silencing of specific neurons , the voxels’ , regions’ , and components’ time series give insight into the dynamics of the network . For example , the widespread activity patterns observed here during walking suggest that coordination between different brain areas is involved . This is just the first step in exploring large-scale brain states , as this technique opens up a new window into the neural mechanisms underlying fly behavior .
We used the GAL4/UAS system to express activity probes in neurons . The fly genotype was as described in the figure legends , and fly stocks were obtained from the Drosophila Bloomington Stock Center , Bloomington , Indiana . Flies were reared at 25 °C with a 24 hour light/dark cycle on brown food ( containing cornmeal , molasses , yeast , soy flour , agar , proprionate , and nipogen ) , which had lower autofluorescence than yellow food ( such as the one from the Bloomington Stock Center , which contains yellow cornmeal , malt extract , corn syrup , yeast , soy flour , agar , and proprionic acid ) . Fly holders were 3D printed using S1 File . A piece of tape ( 0 . 75 inches wide ) was shaped as a 1-mm high step using a 1-mm thick glass slide , and an aperture , as is shown in S1 Fig ( 1 mm wide for the body and 0 . 6 mm wide for the head ) , was made by hand using a sharpened scalpel or a thin lancet ( 36 gauge ) . The tape was then stuck onto the chamber , aligning the opening of the tape to the center of the holder . We added nail polish at the contact between the tape and the holder to avoid leaks . We also added black nail polish to the tape to block the excitation light from hitting the fly’s eyes . Note that although the black painted tape protected the flies’ eyes from direct illumination by the microscope’s excitation light , the light scattered by the brain can still activate the eye’s receptors for blue light , as the transient activity in the first few seconds of each experiment demonstrates ( see , for example , the optic lobe trace in S6 Fig ) . To verify that these receptors were not saturated , we presented flashes of 470-nm blue light as external stimuli ( see S29 Fig ) . Although the stimuli excited fluorophores nonspecifically , PCA and ICA could still extract neuronal calcium responses in the optic lobes and the protocerebral bridge , thus demonstrating that the fly could still perceive external blue stimuli . Besides activation at the onset of excitation light , we observed two types of activity patterns , likely due to the excitation light ( in particular when the ommatidia were not completely protected by the black tape ) but that could also be intrinsic activity ( perhaps important for development ) . First , we observed sudden discharges in medulla columns projecting to lobula layers ( for example , see Fig 7 , second and third traces ) . Second , in some calcium recordings , we observed large oscillating waves propagating onto the medulla and along the lobula . At the start of an experiment , flies were transferred to an empty glass vial and left on ice for approximately 1 minute . The holder was put in contact with wet tissues on ice under a stereomicroscope . A fly from the cold vial was pushed into the holder’s opening so that the posterior part of the head was in contact with the tape . UV-curing glue was added at the junction between the tape and the head between the eyes and cured for 5 seconds using a 365-nm Thorlabs LED light at 20% of power for 5 seconds . A piece of thin copper wire ( wire magnet , 40 gauge ) or a piece of tape was placed above the legs to push them away from the proboscis ( see S1 Fig ) . UV glue was then added at the rim of the eye and all around the proboscis ( which was pushed into the head ) , without touching the antenna or the legs , and was cured for 5 seconds . Uncured glue was carefully removed with tissues . A small amount of vacuum grease was placed around the neck behind the proboscis to avoid later leaks . The wire or tape was then removed , and a small piece of tissue paper or a small polystyrene foam ball was given to the fly to walk on to monitor its health during the following steps . The holder was turned over , and the fly’s thorax was pushed down to clear the way to the back of the brain . Small pieces of tape were added onto any remaining holes around the fly’s body , and UV glue was added on top of them and cured around the thorax to fix it in place . Vacuum grease was then pushed toward the neck with a tissue . Saline ( 108 mM NaCl , 5 mM KCl , 3 mM CaCl2 , 4 mM MgCl2 , 4 mM NaHCO3 , 1 mM NaH2PO4 , 5 mM trehalose , 10mM sucrose , 5 mM HEPES adjusted to pH 7 . 35 +/− 0 . 05 with NaOH , prepared weekly ) was added and left for a few minutes to make sure that there were no leaks . Fresh saline was added , and dissection was started with forceps that had been previously sharpened as finely as possible by hand . We first removed the cuticle in the middle of the back of the head , being careful to cut pieces before pulling them out . This exposed the hole in the middle of the brain where muscle 16 resides . The pulsatile piece was pulled out . Fresh saline was added , and the remainder of the cuticle was removed piece by piece . The brain was washed with saline several times to remove fat bodies . The air sacs were then removed very carefully as to try not to displace the brain . After a new wash with saline , the fly was ready for imaging . The microscope was modified from an upright Olympus BX51W with a 20x NA 1 . 0 XLUMPlanFL or a 40x 0 . 8 NA LUMPLFLN objective ( from Olympus ) . A microlens array with pitch = 125 μm and f/10 to match the 20x objective or f/25 to match the 40x objective [3] ( from RPC Photonics ) was positioned at the image plane using a custom-made holder ( with some parts from Bioimaging Solutions , Inc . ) . Two relay lenses ( 50 mm f/1 . 4 NIKKOR-S Auto from Nikon ) projected the image onto the sensor of a scientific CMOS camera ( Hamamatsu ORCA-Flash 4 . 0 ) . Note that when using half of the camera frame to attain 200 Hz for voltage recordings , the brain fit within the field of view , but rays coming from points far from the focal plane with a large angle were missed , slightly impairing reconstruction . A 490 nm LED ( pE100 CoolLED ) at approximately 10% of its full power was used for excitation . We used a 482/25 bandpass filter , a 495-nm dichroic beam splitter , and a 520/35 bandpass emission filter ( BrightLine , Semrock ) for the fluorescence . We measured the power at the sample with a power meter and found that it was up to 1 mW for the 40x objective and 4 mW for the 20x objective . Photobleaching led to a decrease in intensity after 30 seconds of 13% ( N = 12 , SD = 9% ) for GCaMP6 and 20% ( N = 6 , SD = 13% ) for ArcLight . Note that the full setup cost approximately US$50 , 000 ( US$65 , 000 with the 64 Gb of RAM acquisition computer and the 256 Gb of RAM analysis computers ) , which was substantially cheaper than other cutting-edge microscopy techniques such as two-photon microscopes . The resolution as a function of depth ( see S2 Fig ) was determined by imaging 2-μm fluorescent beads dispersed in an agarose gel . After reconstruction , the center of beads at different distances from the focal plane were recorded using ImageJ , and a MATLAB program measured the point spread function’s axial and lateral full width at half maximum ( see https://github . com/sophie63/FlyLFM for the code ) . The lateral field of view for the 20x objective was 624 x 636 square microns ( 312 x 309 for the 40x objective ) , as was determined using a mire . The fly holder was positioned on a U-shaped stage above an air-supported ball so that the fly could walk ( see Fig 1 ) . The ball was either polyurethane foam ( 10 mm in diameter ) , Styrofoam , or hollow HDPE ( one-fourth inch ) . We prepared a cup matching the ball diameter and with a 1 . 2-mm hole using self-curing rubber ( from Sugru ) or machining aluminum . A bottle of compressed air provided a steady flow in a pipeline consisting of a tube and a pipette tip connected to the cup hole . A micromanipulator ( from Narishige ) positioned the ball under the fly’s legs . For some flies , we instead provided a small Styrofoam ball that the fly could hold . The fly and the ball were illuminated by a row of IR LEDs ( 940 nm ) in front of the fly and were observed at 100 Hz using a small camera ( FFMV-03M2M from Point Grey ) . To better align the behavior with the fluorescence in some experiments , the camera for monitoring the behavior and the fluorescence were synchronized by using the output of the Flash4 . 0 camera to trigger the acquisition from the behavior camera . When imaging fluorescence at 200 Hz , one triggering signal out of two was ignored by the slower behavior camera that recorded at 100 Hz . We recorded the fluorescence images with HCImage ( from Hamamatsu ) and streamed them to RAM on a 64 Gb of RAM computer , which allowed us to record approximately one continuous minute . For the odor stimulus , air was delivered by a pump through an electrically controlled valve ( 12 Vdc normally closed solenoid valve ) , bubbled in 50% ethanol or 50% apple cider vinegar in a vial , and blown toward the fly through an inverted 1-mL pipette tip . The valve circuit was controlled by a relay connected to a LabJack U3-HV through a LJTick-RelayDriver ( from LabJack ) . For visual stimulation , the excitation light and a 365 nm or 470 nm LED were also triggered by the LabJack . The LabJack was controlled using MATLAB programs ( see https://github . com/sophie63/FlyLFM for the code ) . We reconstructed the light field images using a program in Python , as described in [3] . Briefly , a point spread function library corresponding to the specific setup was first generated; we typically chose to reconstruct a stack of 40 layers ( separated by 6 microns ) , with a lateral sampling distance of either 3 or 6 microns . The voltage probe’s low signal-to-noise ratio made it more difficult to detect signals with a finer sampling , so we typically reconstructed the voltage data with a lateral sampling distance of 6 microns and the calcium data with a lateral sampling of 3 microns . We reconstructed the images using 3D deconvolution on a cluster of GPUs ( generously provided by the Qualcomm Institute at UCSD ) . Note that reconstruction using cloud computing ( AWS ) would cost approximately US$0 . 003 dollars per volume . A data set of 10 , 000 time steps required approximately 8 hours to reconstruct on a cluster of 15 GPUs . We assembled the images in a Nifti file using a python routine ( Tiff2niiV2 in https://github . com/sophie63/FlyLFM ) , inspected and cropped them in FIJI [44] , and often discarded the first 5 seconds because the strong activity in response to the excitation light made it difficult to correct for movement and photobleaching . The 3D image registration function 3dvolreg [45] from AFNI was then used to correct for rigid movements . We removed the background fluorescence and the decrease in intensity from photobleaching by subtracting a signal smoothed using a box average over 15 to 30 seconds , depending on the severity of the bleaching and the length of the recording . The time series were then multiplied by −1 for ArcLight data . For denoising , we found that a Kalman filter ( from https://www . mathworks . com/matlabcentral/fileexchange/26334-kalman-filter-for-noisy-movies ) with a gain of 0 . 5 was better than a median filter over 3 points , and we used this for the data in this paper . We then applied SVD to subtract components that were most clearly related to movement; their maps contained shadows around areas with different background intensities as shown in S4 Fig . For some flies , different conditions corresponded to different recordings , which we concatenated in time after preprocessing . The reconstructed data as well as the data after preprocessing is available on the CRCNS website ( https://crcns . org/NWB/Data_sets ) . For early data sets ( before direct synchronization of the cameras ) , the fluorescence and the behavior were aligned using the onset and offset of the excitation light . The small discrepancy ( approximately 30 ms per minute ) between the total time given by the camera for the fluorescence and the camera for the behavior was corrected linearly . The fluorescence data was then interpolated to match the behavior data using the MATLAB function Interpolate2vid . We manually analyzed the behavior ( noting the times of the behaviors or pressing different keyboard keys when we recognized different behavior using the MATLAB GUI Video_Annotate in https://github . com/sophie63/FlyLFM ) . We also characterized the fly’s walk by tracking the movements of the ball using FIJI’s optic flow plugin ( Gaussian Window MSE ) . Maps comparing the activity during rest and walking , resting and grooming , turning left and turning right , and 1 second after stimulus presentation compared to 1 second before were obtained by simply averaging the time series in each voxel for the different conditions and subtracting these maps from one another . The positive value was colored in magenta and the negative in green , thus showing which condition dominated in which voxel . The average of the volume time series was aligned to an anatomical template ( available from https://github . com/VirtualFlyBrain/DrosAdultBRAINdomains ) in which the brain is segmented into regions according to the nomenclature described in [14] . The registration was performed using landmarks with ImageJ ( as described in http://imagej . net/Name_Landmarks_and_Register ) . We marked several points in the protocerebral bridge , the tips of the mushroom body ( between the alpha and alpha’ lobes ) , the middle of the noduli , the lateral tip of the lateral triangles , the lateral tip of the lateral horns , the center of the ellipsoid boy , the center of the antennal lobes , and the bottom part of the trachea at the level of the noduli . Although the landmarks were readily observable with the background fluorescence ( see S3 Fig , for example ) , making a template superposing the components to the volume average helped to visually find the landmarks . Dimensionality was then reduced by separating the volumes into slices of thickness corresponding to the point spread function height and averaging in z for each slice . The 4D data sets were typically ( x , y , z , t ) = 200 x 100 x 10 x 10 , 000 at this stage . For source extraction , we found that using melodic [12] from the FSL package readily produced meaningful components . However , as the code was running slowly on our large data sets , we adapted it in MATLAB to parallelize some steps . A first step of SVD was used to remove the largest part of the components before calculating the pixel-wise temporal variance ( without spatial filtering ) that was used to normalize the data . We then reapplied SVD and plotted the log of the singular value spectrum to automatically detect the shoulder at the point with a 45° tangent . We found that although the components with the smallest variance were noise , some activity-related components were still present after the shoulder point . As such , choosing twice the number of components at the shoulder gave a good compromise between keeping activity and removing noise components . ICA was then applied to the SVD spatial components with FastICA [46] ( see ICAalamelodic . m file from https://github . com/sophie63/FlyLFM ) . The sign was chosen so that the average of the positive side of the map was larger than the negative side . The components were then automatically sorted by brain region; after registering the standard brain to the data , we averaged the components maps in anatomically defined regions [14] using regions masks and chose the main region as the one with the highest average . We removed components corresponding to movement or noise partly automatically ( removing components present in more than five regions and containing more than 200 separate objects ) and partly by hand ( see example of typical artifactual components in S11 Fig ) using a Jupyter notebook ( the notebooks corresponding to the choices made for the figures in this paper can be found at https://github . com/sophie63/FlyLFM ) . To obtain the time series from regions of interest , we first made masks using the PCA/ICA maps . We calculated the standard deviation from the value in the map and set all voxels with values inferior to three times that standard deviation to zero . We then used those masks to do a weighted average of the ΔF/F time series . The time series for turning left or right were obtained by convolving the optic flow for the ball going left or right , with a kernel corresponding to the GCaMP6 impulse response . These time series were then regressed with the components’ time series , and we inspected the maps with the strongest regression coefficients . Fig 1B bar was added with ImageJ , and the 3D rendering was done in Icy [47] , in which transparency and contrast were adjusted globally on the volume . The component’s maps were thresholded at 3x standard deviation , and only the positive part of the maps was displayed . The image contrast was then globally adjusted in ImageJ , and the figures panels were assembled in Inkscape . The MATLAB and Python code for preprocessing , PCA/ICA , and sorting of the components is available at https://github . com/sophie63/FlyLFM .
|
Whole-brain recordings give us a global perspective of the brain in action . This is already possible in humans , for which functional magnetic resonance imaging ( fMRI ) has opened a new chapter in the study of brain activity underlying behavior , but this technique has low spatial and temporal resolution . In animals , techniques for imaging a whole brain so far have allowed us to record activity at much higher spatial resolution , but these are still orders-of-magnitude slower than neuronal electrical activity . Here , we have developed a technique for ultra-fast imaging of whole-brain activity in fruit flies while they are behaving ( walking , resting , or grooming ) and when they perceive various stimuli . We find that there is a global increase in activity when the fly walks compared to when it rests , while only a small local increase is observed when the fly grooms compared to when it rests . We have also used computational techniques to extract activity from small brain regions or from specific neuron types and identified regions involved in turning left or right as well as regions with ongoing activity in the absence of stimuli or behavior .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"methods",
"and",
"resources",
"statistics",
"light",
"neuroscience",
"electromagnetic",
"radiation",
"multivariate",
"analysis",
"mathematics",
"artificial",
"light",
"computational",
"neuroscience",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"optic",
"lobes",
"imaging",
"techniques",
"animal",
"cells",
"mathematical",
"and",
"statistical",
"techniques",
"principal",
"component",
"analysis",
"neuropil",
"glial",
"cells",
"physics",
"cellular",
"neuroscience",
"cell",
"biology",
"anatomy",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"computational",
"biology",
"ocular",
"system",
"statistical",
"methods"
] |
2019
|
Fast near-whole–brain imaging in adult Drosophila during responses to stimuli and behavior
|
Learning to read is a fundamental developmental milestone , and achieving reading competency has lifelong consequences . Although literacy development proceeds smoothly for many children , a subset struggle with this learning process , creating a need to identify reliable biomarkers of a child’s future literacy that could facilitate early diagnosis and access to crucial early interventions . Neural markers of reading skills have been identified in school-aged children and adults; many pertain to the precision of information processing in noise , but it is unknown whether these markers are present in pre-reading children . Here , in a series of experiments in 112 children ( ages 3–14 y ) , we show brain–behavior relationships between the integrity of the neural coding of speech in noise and phonology . We harness these findings into a predictive model of preliteracy , revealing that a 30-min neurophysiological assessment predicts performance on multiple pre-reading tests and , one year later , predicts preschoolers’ performance across multiple domains of emergent literacy . This same neural coding model predicts literacy and diagnosis of a learning disability in school-aged children . These findings offer new insight into the biological constraints on preliteracy during early childhood , suggesting that neural processing of consonants in noise is fundamental for language and reading development . Pragmatically , these findings open doors to early identification of children at risk for language learning problems; this early identification may in turn facilitate access to early interventions that could prevent a life spent struggling to read .
Three aspects of auditory-neurophysiological processing have often been associated with literacy: variability of neural firing [1 , 2] , auditory system timing [3 , 4] , and processing detailed acoustic features such as those found in consonants [5 , 6] . This neural coding is thought to play a pivotal role in reading and language development [5 , 7 , 8] and may reflect the precision of neural processing in the central auditory system , which likely develops through the integrated neural coding of speech across multiple timescales , including prosodic , syllabic , and phonemic acoustic information [8–10] . Although children are provided access to these sonic fundamentals in their everyday lives , these experiences often occur in adverse listening environments ( classrooms , outdoors , wailing siblings ) in which children need to tune out competing sounds to tune into speech . Indeed , noise places stringent demands on sensory processing , and individuals with language-based learning problems often have perceptual deficits in noise across modalities [11–15] . Background noise limits access to redundant acoustic cues that are accessible to listeners in quiet . In principle , noise may obfuscate both the neural processing of an individual acoustic event ( such as a phoneme ) and the formation of consistent representations of successive events ( such as words or sentences ) ; see , for example , [16] . Should children with poor processing in noise grow up forced to make sense of speech in these noisy environments , they may fall behind their peers in language development . Auditory system precision—especially the neural processing of speech in noise—is correlated to literacy; that is , struggling readers perform poorly on behavioral tests of auditory processing [4] and have reduced auditory response fidelity and impaired neural coding of rapid auditory stimuli compared to good readers [2 , 17] . Therefore , these brain–behavior links likely reflect neural mechanisms underlying reading in general , as opposed to a parochial deficit in clinical populations . It remains open to debate , however , what role these neural mechanisms play developmentally with respect to reading , in part because it remains debated if auditory function is consistently implicated in reading impairment at all [18] . Alternate accounts for the origins of reading impairment include sluggish processing in the magnocellular pathway [19 , 20] , multimodal perceptual deficits grounded in inefficient short-term memory [21] , and poor processing in cortical “reading networks” that lead to auditory impairments [22] . There are likely many reasons that a child may be a poor reader , including genetic and environmental; while understanding the factors that cause reading impairment is an important goal , it is also important to predict which children will struggle when they begin to read . Thus , from a pragmatic standpoint our aim is to define a neurophysiological marker that might identify these children . To date , auditory-neurophysiological markers of literacy have only been observed in children and adults who have received prolonged , formal instruction . But the process of learning to read itself may induce changes in substrate reading skills [23 , 24] and their neural foundations [25] . Further compounding the problem is the challenge of predicting future literacy skills . There have been promising experiments reporting differences between groups of children ( e . g . , an at-risk group versus a control group or a group of children who receive a diagnosis versus a group who does not ) . But substantial overlap between groups ( resulting in modest effect sizes ) tends to thwart clinically-meaningful predictions in individual children [26–28] . Early identification of children at risk for reading problems is crucial; interventions that are provided early enough can bring struggling pre-readers in line with their peers and offset years of reading difficulties [29 , 30] . For example , in a prospective study of language-impaired children , Bishop and Adams reported that literacy development proceeded smoothly in children whose oral language problems were resolved by age 5 . 5 y [31] . This motivates us to investigate early language skills , and their neural correlates , in preschoolers . It has long been argued that reading skills are linked to the processing of rapid auditory information , meaning that struggling readers have particular problems with auditory temporal processing [4 , 5 , 32] , including the perception and neural coding of dynamic speech elements [11 , 15] . Here , then , we evaluated neural processing of a consonant-vowel syllable in background noise . This processing in noise relies upon neural synchrony—that is , consistent and uniform neural population discharges [33] . In humans , neural synchrony in response to the crucial phonemic features of speech may be measured through the frequency following response ( FFR , a scalp-recorded auditory evoked potential that is also known as the auditory brainstem response to complex sounds , or cABR ) . The neural circuitry important for language development may not engage faithfully during everyday listening experiences because of a breakdown in synchronous neural firing exacerbated by background noise . As a consequence of this poor online processing in noise , these children may lag behind their peers in language development . Previous studies in older children have established relationships between FFR properties and reading , and therefore provide empirical grounding for the current investigation [2 , 3 , 11] . We also evaluated children’s phonological skills because phonological processing—knowledge and manipulation of the sound structure of spoken language—is a chief pre-reading skill that is deficient in children with dyslexia [8] . Our hypothesis is that background noise disrupts brain mechanisms involved in literacy development; we therefore predict that children with poor auditory-neurophysiological responses to speech in noise exhibit poorer early literacy skills than their peers .
We constructed a statistical model incorporating three aspects of the neural coding of consonants in noise: trial-by-trial stability [1 , 2] , neural timing [3 , 15] , and representation of spectral features that convey phonemic identity ( see Fig 1 ) [3 , 11] in a cohort of 4-y-old children who had not yet learned to read ( n = 37 , 21 female; mean [M] age 54 . 41 months , standard deviation [SD] 3 . 56 ) . These quantify different aspects of auditory processing and have all been linked to reading skills in older children . Although these metrics come from a single neurophysiological recording , they are not strongly intercorrelated within an individual ( see S1 Fig ) ; thus , each provides unique information about the coding of different linguistic and paralinguistic parameters . We found that neural coding of consonants in noise strongly predicted phonological processing in pre-readers over and above demographic factors ( CELF P-2 Phonological Awareness; ΔR2 = 0 . 488 , F[9 , 24] = 4 . 121 , p = 0 . 003; total R2 = 0 . 684 , F[12 , 36] = 4 . 328 , p = 0 . 001; see Table 1 and Fig 2A; when the correlation was adjusted for test-retest variability of the behavioral test , R2 = 0 . 757; see also S1 Text for a cross-validation of this model ) . For the majority of children , our model predicted scores within 2 points on the test , which is less than a 10% margin of error ( difference between actual scores and model-predicted scores; median = 1 . 97 points; range , 0 . 17–5 . 66 points; see Fig 2B ) . Our results suggest that the precision and stability of coding consonants in noise parallels emergent literacy skills across a broad spectrum of competencies—all before explicit reading instruction begins . Statistical model predictions from this regression were used in subsequent analyses . The idea is that model predictions reflect a “consonants-in-noise score” that may be correlated to performance cross-sectionally and longitudinally on additional behavioral tests . For Experiments 2 through 4 , we measured FFRs to consonants in noise , computed the same measures of neural coding in those children , and applied regression parameters from Experiment 1 to those children’s responses . This effectively predicts performance on this test of phonological processing even though , as detailed below , we did not conduct this particular test in all children . In no cases did we refit the data with new regression models . Having constructed a model based on phonological processing , we explored whether model predictions generalized to multiple tests of preliteracy . We applied our predictive model from Experiment 1 to 20 3-y-olds ( 9 female; M = 43 . 35 months , SD 2 . 50 ) in whom we could not administer the test of phonological processing ( see Methods ) but could conduct neurophysiological testing . We used the model parameters estimated in Experiment 1 and combined these “consonants-in-noise scores” with those from the 37 children in that experiment . Neural coding of consonants in noise predicted performance on a test of rapid automatized naming , an additional key preliteracy skill that is thought to be highly predictive of future reading success across languages [34 , 35] ( higher predicted scores correlated with faster naming; r[55] = -0 . 550 , p < . 001 ) . Neural coding also predicted children’s memory for spoken sentences ( r[55] = 0 . 516 , p < . 001 ) , a test that combines auditory working memory with knowledge of grammar—an additional substrate skill that contributes to literacy development and is often deficient in children with dyslexia and/or language impairment [36] . We also split this cohort into the two age groups . Recall that the “consonants-in-noise score” was fit to the 37 4-y-olds from Experiment 1 , and we applied these regression weights to the 20 3-y-olds in whom we could not measure phonological processing . In the 4-y-olds the “consonants-in-noise” score predicted memory for spoken sentences ( r[35] = 0 . 555 , p < . 001 ) and trended towards predicting faster rapid naming ( r[35] = -0 . 301 , p = . 070 ) . Crucially , in the 3-y-olds the model predicted rapid naming ( r[18] = -0 . 692 , p = . 001 ) , meaning that applying the model derived in Experiment 1 generalizes both to a new cohort and a new preliteracy skill; however , it did not predict 3-y-old’s memory for spoken sentences ( r[18] = 0 . 034 , p = 0 . 888 ) . Scatterplots for these correlations are shown in S3 Fig . A subset of children from Experiments 1 and 2 returned one year later for a behavioral test battery ( n = 34 , 18 female ) . We took the “consonants-in-noise score” derived from the model in Experiment 1 and explored relations between the model’s predictions and performance on a variety of literacy tests one year after neurophysiological assessment . Year 1 neurophysiological testing predicted future performance on the same test of phonological processing—including in children too young to take this test in Year 1 ( r[32] = 0 . 543 , p = . 001 ) . These predictions generalized to future performance on a second test of phonological processing ( r[32] = 0 . 575 , p < . 001 ) and predicted future performance on the same test of rapid automatized naming ( r[32] = -0 . 663 , p < . 001; see Fig 3 ) and the same test of memory for spoken sentences ( r[32] = 0 . 458 , p = 0 . 006 ) . In the second year we also administered tests to evaluate early literacy . Neurophysiological model predictions at Year 1 predicted future performance on sight word reading ( r[32] = 0 . 476 , p = . 004 ) , spelling ( r[32] = 0 . 415 , p = . 015 ) , and a composite reading score ( r[32] = 0 . 425 , p = . 012; see S4 Fig ) . Thus , the neural coding of consonants in noise predicts future reading achievement on standardized tests , in addition to multiple substrate literacy skills . In Experiments 1–3 , we established an auditory-neurophysiological biomarker for pre-reading skills in preschoolers . We applied the regression model from Experiment 1 to a cohort of older children ( n = 55 , 22 female , ages 8–14 y , M = 10 . 82 , SD = 1 . 7 ) in whom we collected identical auditory-neurophysiological responses ( previously described in [15] ) . This allowed us to ask whether the “consonants-in-noise score” derived in the 4-y-old children generalizes to a different age group , and effectively predicts how these children would have performed on the preschool tests of phonological processing , given their precision of coding consonants in noise . In school-aged children , the neural coding of consonants in noise predicted concurrent reading competence ( r[53] = 0 . 430 , p = . 001 ) and performance on a range of literacy tests including sight word reading ( r[53] = 0 . 408 , p = . 002 ) , non-word reading ( r[53] = 0 . 329 , p = . 014 ) , spelling ( r[53] = 0 . 327 , p = . 015 ) , oral reading efficiency ( r[53] = 0 . 319 , p = . 018 ) , and phonological processing ( r[53] = 0 . 474 , p < . 001; see S5 Fig ) . A subset of these children had been externally diagnosed with a learning disability ( LD; n = 26 ) ; the diagnostic groups differed on their predicted scores ( F[1 , 53] = 14 . 541 , p < . 001 ) and model predictions reliably classified children into diagnostic categories ( discriminant function analysis: 69 . 1% of cases correctly classified , λ = 0 . 785 , χ2 = 12 . 728 , p < . 001 ) . A receiver operating characteristic ( ROC ) analysis ( see S6 Fig ) revealed that the model score excelled in identifying if a child was not in the reading-impaired group ( area under the curve [AUC] = 0 . 756; 95% confidence interval [CI] , 0 . 627 , 0 . 885; p = . 001 ) . From a clinical standpoint , this suggests that our consonants-in-noise approach may be most effective in “clearing” children as unlikely to develop an LD , thereby motivating thorough follow-up in the remaining children .
A well-acknowledged gap in our understanding of the biology of reading is what biological constraints are instantiated in the nervous system prior to reading instruction . Ours is , to our knowledge , one of the first studies to demonstrate a physiological–phonological coupling in an age group sufficiently young to preclude confounds from prolonged and formal reading experience . In this respect our findings are consistent with the view that phonological processing is a necessary foundational skill for reading development [8 , 24] . By establishing brain–behavior links in pre-readers that are carried through to school-aged children , our findings suggest a causal , and not simply correlative , role for auditory processing in learning to read . Because the integrity of neural speech processing is linked to phonological awareness ( to date , perhaps the best conventional predictor of a child’s eventual reading achievement [37] ) we suggest that the neurophysiological markers we report here provide a biological looking glass into a child’s future literacy . Indeed , we show that our model predicts performance on reading readiness tests one year after neurophysiological assessment . In many cases , behavioral tests were not standardized for children as young as we could evaluate neurophysiologically . Moreover , we show that , in school-aged children , our model predicts literacy and diagnostic category . Thus , in cases of learning disabilities , this biomarker may represent pre-existing problems with forming sound-to-meaning and/or letter-to-sound connections that cause problems for children when they begin reading instruction , an interpretation in line with converging biological evidence [27 , 38] . The correlations between neural coding and literacy skills were somewhat weaker in school-aged children than in pre-readers; this is consistent with the view that reading subskills mature as a function of reading experience , and that phonological processing may not play as strong a role in literacy competence for older children as it does during the early stages of reading acquisition [39 , 40] . Moreover , older children may have developed compensatory strategies that reduce the influence of phonological processing on reading that contributed to this developmental uncoupling . Nevertheless , it is noteworthy that there was a consistent brain–behavior relationship observed from ages 3–14 . Taken together with the breadth of relationships observed across preliteracy skills ( i . e . , both phonological processing and rapid naming ) , the neural coding of consonants in noise may reflect a child’s core literacy potential . Pharmacological studies have suggested that the neurophysiological metrics in our model rely on inhibitory neurotransmitter function; a loss of inhibitory receptors and/or an excitatory-inhibitory misbalance in auditory midbrain is linked directly to a decrease in the synchronous neural firing necessary to encode dynamic speech features such as consonants [41] , especially in adverse listening conditions . In fact , this subcortical neural synchrony is necessary for auditory processing in noise [33] . We therefore speculate that the biomarker revealed here may rely on the emergence of robust inhibitory function . By measuring suprathreshold responses to consonants in noise , we may have sufficiently taxed the developing auditory brain to reveal systematic individual differences in inhibitory processing . Individual differences in these functions may create challenges when children are trying to map sounds to meaning in noisy environments , potentially interfering with the development of the range of preliteracy skills correlated to auditory-neurophysiological responses here . Our view is that this subcortical neural synchrony emerges and is honed through a distributed , but integrated , auditory circuit . With respect to reading , auditory cortical processing is thought to bootstrap the development of fluent speech processing; eventually , children begin to associate orthographic representations with mental representations of phonemes [8 , 10 , 17] . A breakdown in this integrative process may cause a reduction in corticofugal input in auditory midbrain ( our biomarker’s putative generator ) , especially for acoustic transients in challenging listening environments ( i . e . , consonants in noise ) . This faulty processing may be due to poor phaselocking [10] , abnormal thalamic and cortical cytoarchitectonics [38 , 40 , 42–44] , and/or sluggish attentional resources [45] . Should a child fail to learn what to pay attention to in everyday listening environments , and in turn fail to allocate appropriate attentional resources to these relevant speech cues , he or she may struggle to build robust phonemic representations . This sound-meaning disjunction may disrupt the course of auditory learning , leading to suboptimal input from corticocollicular fibers and cascading to a decrease in inhibitory function at the cost of synchronous firing by midbrain nuclei [41] . In turn , without the development of refined neural coding , maladaptive compensatory mechanisms may develop that stanch the development of automaticity in reading and auditory processing in a feed-forward , feed-back loop . This view is consistent with evidence that substrate reading skills ( such as phonological processing ) and sensory processing develop as a function of reading experience [25 , 46] . Of course , this is speculative; we must infer midbrain function from far-field electrophysiological recordings . Nevertheless , it is intriguing to contemplate the role of inhibitory neurotransmission , and neurochemical mechanisms more broadly , with respect to language development [47] . Conventional tests of early literacy can be unreliable in children this young , and to our knowledge , standardized tests of phonological processing are not available for children younger than age 4 . Moreover , children who perform poorly on these tests have the least reliable scores because the fewest items are administered , thereby increasing potential bias from a false positive . Given the comorbidity between reading disorders and other LDs , compliance with paper-and-pencil tests may be even lower in the children who stand at the highest risk for a disability and are the most important cases to screen . When these evaluations are available , they are most reliable in identifying a child at risk for a LD , rather than systematically predicting a child’s position along a continuum of literacy achievement . The same may be said for previously established neurophysiological predictors of a child’s diagnosis [28 , 48] . We do not make these claims to denigrate the contributions of other research groups , or the obvious fact that , in many cases , simple paper-and-pencil tests and surveys can be effective in evaluating a child’s risk for a learning problem . Rather , our view is that by establishing these brain–behavior links in preschool children , our findings can pave the way for auditory-neurophysiological assessment in even younger children , in addition to children who are difficult to test using conventional means . Our approach was to combine multiple measures of neural coding to see how they collectively predict preliteracy skills; although all came from the same neurophysiological recording , each provided unique information and they were only modestly intercorrelated ( average r = 0 . 318 ) . Future work should focus on the similarities and differences between these measures . On the one hand , we provide evidence that in combination they predict several preliteracy skills and diagnostic category . On the other hand , reading impairment can arise for a number of reasons , which may have distinct pathophysiologies [49] . An intriguing possibility is that these different aspects of neural coding are uniquely linked to different etiologies of reading impairment and/or substrate reading skills . These children will continue to be followed longitudinally to better understand the role this neural coding in noise plays in language development . From a theoretical perspective , we hope to elucidate how consonant processing in noise guides the development of literacy skills , especially in interactions with the distributed-but-integrated neural networks involved in auditory learning . Children with particularly poor processing of speech in noise may face challenges during critical auditory mapping experiences [50] , inhibiting the development of precise neural coding . It would appear that we have established a neural correlate of preliteracy that is carried through to school age , precedes explicit reading instruction , and predicts both a child’s performance along a continuum of literacy and diagnostic category; it will be necessary , however , to replicate these findings in a larger sample . Pragmatically , our findings have the potential to facilitate both early diagnosis and interventions to improve literacy before a child begins explicit instruction . Efforts to promote literacy during early childhood can be tremendously effective , and our hope is that these results open a new avenue of early identification to provide children access to these crucial interventions .
Children were recruited from the Chicago area . No child had a history of a neurologic condition , diagnosis of autism spectrum disorder , or second language experience ( all were native English speakers ) . In all cases children had normal auditory brainstem responses ( elicited by a 100 μs square-wave click presented at 80 dB SPL to the right ear at 31 . 3 Hz; Navigator Pro , Bio-Logic Systems , Mundelein , IL , United States ) . Preschoolers ( Experiments 1–3 ) passed a screening of peripheral auditory function ( normal otoscopy , Type A tympanograms , distortion product otoacoustic emissions ≥ 6 dB SPL above the noise floor from 0 . 5–4 kHz ) . School-aged children ( Experiment 4 ) passed an audiometric screening ( air-conduction thresholds ≤15 dB HL at octaves from 0 . 250–8 kHz bilaterally with no evidence of a conductive hearing loss and distortion product otoacoustic emissions ≥6 dB SPL above the noise floor from 0 . 5–4 kHz ) . Frequency-following responses were elicited to a 170 ms [da] stimulus . The [da] is a voiced ( 5 ms voice onset time ) six-formant stop consonant constructed in a Klatt-based synthesizer at 20 kHz . Following the initial stop burst is a 50 ms consonant transition ( /d/ to /a/ ) during which the lower three formants shift in frequency ( F1 400–720 Hz , F2 1 , 700–1 , 240 Hz , F3 2 , 580–2 , 500 Hz ) ; these formants are steady for the subsequent 120 ms vowel ( /a/ ) . The fundamental frequency and upper three formants are steady throughout the stimulus ( F0 100 Hz , F4 3 , 300 Hz , F5 3 , 750 Hz , F6 4 , 900 Hz ) . The stimulus was presented against a six-talker babble track at a +10 SNR . The babble track consists of six talkers ( three female ) speaking semantically-anomalous English sentences . The 4 , 000 ms babble track is looped continuously such that there is no phase synchrony between the onsets of the [da] and noise . The [da] and noise were mixed into a single channel that was presented to the right ear at 80 dB SPL in alternating polarities through electromagnetically-shielded insert earphones ( ER-3A , Etymotic Research , Elk Grove Village , IL , US ) . Children sat in an electrically shielded and sound-attenuated booth ( IAC Acoustics , Bronx , NY , US ) and sat in a comfortable chair for recording while watching a film of their choice . The left ear remained unoccluded so the children could hear the movie soundtrack ( ~40 dB SPL ) . Our selection of metrics from the FFRs was motivated by previous investigations that have found links cross-sectionally between the timing , stability , and magnitude of responses to consonants and literacy skills . By using the same stimulus and recording scheme , we can apply uniform neurophysiological analyses across age groups . Please see [52] for technical guidance on FFR collection and analysis . A series of standardized psychoeducational tests were administered . As much as possible , these tests were selected to provide overlap between experiments; however , we were constrained by the ages for which the tests were standardized and available . Please see Table 2 for a summary of each behavioral test broken down by experiment . The test battery included the Children’s Evaluation of Language Fundamentals-Preschool 2nd Edition ( CELF-P2; Phonological Awareness and Recalling Sentences subtests; raw scores; Pearson , San Antonio , TX , US ) , the RAN ( rapid automatized color and object naming; average naming time in seconds normalized on a log scale; PRO-ED , Inc . , Austin , TX , US ) , the Comprehensive Test of Phonological Processing ( CTOPP; 1st Edition for school-age children , 2nd Edition used for preschoolers; composite phonological awareness score used , standard score; Pearson , San Antonio , TX , US ) , the Woodcock-Johnson-III Tests of Achievement ( WJ-III; Letter-Word Identification , Spelling , and Word Attack subtests and Basic Reading composite , standard scores; Riverside Publishing , Rolling Meadows , IL , US ) and the Test of Word Reading Efficiency ( TOWRE , standard scores; Pearson , San Antonio , TX , US ) . Non-verbal intelligence was evaluated in preschoolers with the Wechsler Preschool and Primary Scale of Intelligence-III ( WPSSI-III , Object Assembly in 3-y-olds and Matrix Reasoning in 4-y-olds; scale scores; Pearson , San Antonio , TX , US ) and in school-age children with the Wechsler Abbreviated Scale of Intelligence ( WASI , Matrix Reasoning and Block Design subtests , standard scores; Pearson , San Antonio , TX , US ) . Hierarchical regression was used to predict phonological processing from neurophysiological recordings . The first step comprised demographic factors ( age , sex , and non-verbal intelligence ) and the second step comprised neurophysiological factors; thus , the model estimates what percentage of variance in phonological processing neural coding accounts for above and beyond demographics . The model constructed in Experiment 1 was applied to all subjects; on its first step there was a trend for demographics to significantly predict phonological processing ( R2 = 0 . 183 , F[3 , 37] = 2 . 547 , p = 0 . 072 ) . In preliminary modeling , independent two-step regressions were run for each neurophysiological metric . In all cases , the neurophysiological metrics in isolation improved model fit ( neural timing: ΔR2 = 0 . 245 , F[4 , 29] = 3 . 166 , p = 0 . 028; representation of first formant: ΔR2 = 0 . 254 , F[4 , 29] = 3 . 340 , p = 0 . 023; neural stability: ΔR2 = 0 . 142 , F[1 , 32] = 6 . 166 , p = 0 . 013 ) . These regression results are presented in S2 Table as Steps 2A , 2B , and 2C , respectively . Despite these metrics coming from a single recording , the overall model had acceptable levels of collinearity ( tolerance ranged from 0 . 383–0 . 994 ) , indicating that the model was not skewed by intercorrelations between predictors . All variables met the assumptions of the general linear model ( i . e . , normal distribution and heterogeneity of variance ) and p-values reflect two-tailed tests .
|
Learning to read is a chief developmental milestone with lifelong consequences; although there are effective interventions for struggling readers , an ongoing challenge has been to identify candidates for intervention at a young-enough age . We measured the precision of the neural coding of consonants in noise , and found that pre-reading children ( 4 y old ) with stronger neural processing had superior early literacy skills; one year later they were also stronger emerging readers . We applied the same neural coding measure to a cohort of older children: in addition to predicting these children’s literacy achievement , we could reliably predict which of the children had received a diagnosis of a reading impairment . Taken together , these results suggest that the neural coding of speech in noise plays a fundamental role in language development . Children who struggle to listen in noisy environments may struggle to make meaning of the language they hear on a daily basis , which can in turn set them at risk for literacy challenges . Evaluating the neural coding of speech in noise may provide an objective neurophysiological marker for these at-risk children , opening a door to early and specific interventions that may stave off a life spent struggling to read .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Auditory Processing in Noise: A Preschool Biomarker for Literacy
|
Gallium is a semi-metallic element known since the 1930s to have antimicrobial activity . This activity stems primarily from gallium's ability to mimic trivalent iron and disrupt specific Fe ( III ) -dependent pathways , particularly DNA synthesis ( due to inhibition of ribonucleotide reductase ) . Because of its novel mechanism of action , gallium is currently being investigated as a new antibacterial agent , particularly in light of the increasing resistance of many pathogenic bacteria to existing antibiotics . Gallium maltolate ( GaM ) is being developed as an orally and topically administrable form of gallium . Yaws is a neglected tropical disease affecting mainly the skin and skeletal system of children in underprivileged settings . It is currently the object of a WHO-promoted eradication campaign using mass administration of the macrolide azithromycin , an antibiotic to which the yaws agent Treponema pallidum subsp . pertenue has slowly begun to develop genetic resistance . Because yaws transmission is mainly due to direct skin contact with an infectious skin lesion , we evaluated the treponemicidal activity of GaM applied topically to skin lesions in a rabbit model of yaws . Treatment efficacy was evaluated by measuring lesion diameter , treponemal burden in lesion aspirates as determined by dark field microscopy and amplification of treponemal RNA , serology , and immunohistochemistry of biopsied tissue samples . Our results show that topical GaM was effective in reducing treponemal burden in yaws experimental lesions , particularly when applied at the first sign of lesion appearance but , as expected , did not prevent pathogen dissemination . Early administration of GaM to yaws lesions could reduce the infectivity of the lesions and thus yaws transmission , potentially contributing to current and future yaws control campaigns .
Yaws is a neglected tropical disease ( NTD ) [1 , 2] caused by the spirochete bacterium Treponema pallidum subsp . pertenue ( Tp pertenue ) , a pathogen closely related to the syphilis agent , Treponema pallidum subsp . pallidum ( Tp pallidum ) [3] . Yaws mainly affects children less than 15 years of age , among whom the disease is spread via skin contact with an infectious early lesion [4 , 5] , although a marginal role in transmission may be played by vector flies and non-human primates [6 , 7] . Similar to syphilis , untreated yaws becomes a multistage chronic disease that mainly affects the skin and skeletal system of infected individuals [2 , 8–11] . In contrast to syphilis , yaws is believed not to affect either the cardiovascular system or the central nervous system ( CNS ) , and not to be vertically transmitted , even though several studies suggest that CNS , cardiovascular , and fetal involvement cannot be ruled out [12–15] . A detailed review of the early and late clinical manifestations of this disease is available elsewhere [2 , 8–11 , 16–20] . Yaws is currently reported from 14 countries in the Western Pacific , South-East Asia , and African WHO regions where , collectively , about 65 , 000 new cases of yaws per year have occurred since 2008 [21] . The highest disease incidence is reported in Papua New Guinea , the Solomon Islands , and Ghana [22–26] . A major anti-yaws campaign in the 1950s and 1960s by the WHO and UNICEF eradicated about 95% of the disease in 46 developing countries , causing its prevalence to drop from 50 million cases ( reported in 1952 ) to 2 . 5 million cases ( reported in 1964 ) [27] . This success induced the WHO to gradually eliminate its eradication programs , confident that the primary healthcare facilities established during the campaign would identify and eliminate the remaining cases . Lack of commitment and resources , however , led to disease resurgence in several countries [17] . In 1995 , the yaws global prevalence in children was estimated to be of approximately 500 , 000 cases [21] . In 2013 , a new campaign to achieve global eradication of yaws by 2020 was initiated by the WHO [28] . This new effort was warranted by the evidence that a single oral dose of azithromycin proved as effective as injected benzathine penicillin in curing yaws [29] . Using azithromycin could avoid the intrinsic difficulties associated with the use of penicillin , which requires an efficient cold chain and personnel able to perform injections . The use of azithromycin , however , has induced the insurgence of macrolide-resistant yaws strains whose spread could undermine the success of the ongoing campaign [30] . Additionally , even when treated with systemic antibiotics like azithromycin or penicillin , lesions might remain contagious for several hours to days post-treatment , based on studies of drug administration in the rabbit model of syphilis [31] . In this context , the application of a topical anti-treponemal agent unable to induce genetic resistance could be useful to reduce transmissibility . Gallium ( Ga ) is a semi-metal that has been extensively studied as an anticancer agent , and is currently being evaluated for repurposing as a novel antimicrobial agent due to its demonstrated activity against pathogenic bacteria and its very low human toxicity [32–38] . In the early 1930s , prior to the discovery of penicillin , experiments conducted at the Pasteur Institute in Paris , France , supported the efficacy of some gallium compounds , particularly “gallium tartrate” ( GaT ) , against Tp pallidum and trypanosomes [39] . These studies claimed that administration of GaT eradicated treponemes from several infected rabbits within three to four days after a single intravenous or intramuscular injection and caused the then-used Meinike reaction ( a serum-induced precipitation of cholesterolized organ extracts performed to diagnose an active Tp infection as an alternative to the modern non-treponemal tests ) to become negative [39] . The antimicrobial activity of Ga is due primarily to it acting as a non-functional mimic of Fe ( III ) [34 , 40] . Unlike Fe , which readily cycles between trivalent and divalent states , Ga is not reducible under physiologic conditions , remaining as Ga ( III ) . By competing with Fe ( III ) , Ga ( III ) can inhibit many Fe ( III ) -dependent biochemical activities , the most prominent being the activity of ribonucleotide reductase to synthesize DNA [34] . Recent interest in Ga compounds as antimicrobial agents [41 , 42] has been motivated by the need for new approaches to fight antibiotic-resistant bacteria and by the shortage of new antibiotics in the pharmaceutical pipeline . Gallium maltolate ( GaM ) is currently under investigation as an orally and topically administrable form of Ga [38 , 40] . GaM is pH and charge neutral , and is moderately soluble in both water and lipids , making it well suited for pharmaceutical administration [40] . Locally administered GaM was effective against Pseudomonas aeruginosa in a mouse burn/infection model [43] , it was also effective against Staphylococcus aureus and methicillin-resistant S . aureus ( MRSA ) [44] and several veterinary pathogens [45–49] . Additionally , topical GaM provided in a water/hydrophilic petrolatum emulsion was shown to have anti-inflammatory and analgesic activity in people with neuropathic pain and inflammatory conditions [38 , 50–53] . Here , we investigated the efficacy of topical GaM against Tp pertenue in a rabbit model of yaws .
New Zealand White ( NZW ) rabbits were used for propagation of Tp pertenue and intradermal ( ID ) experimental infections to assess efficacy of topical GaM . Animal care was provided in accordance with the procedures described in the Guide for the Care and Use of Laboratory Animals [54] under protocols approved by the University of Washington Institutional Animal Care and Use Committee ( IACUC ) . The protocol number assigned by the IACUC committee that approved this study is 4142–01 . No investigations using human samples or humans were conducted in this study . Outbred adult male NZW rabbits ranging from 3 . 5–4 . 5 Kg were purchased from Western Oregon Rabbit Co . ( Philomath , OR ) . Rabbits were housed at 16°C to 18°C in individual cages and fed antibiotic-free food and water . Prior to entry into the study , to rule out previous infection with the rabbit syphilis agent Treponema paraluiscuniculi , each animal was bled and heat-inactivated sera were tested individually with both the fluorescent treponemal antibody absorption ( FTA-ABS ) and Venereal Disease Research Laboratory ( VDRL; BD , Franklin Lakes , NJ ) tests according to the manufacturer’s instructions . Only rabbits seronegative to both tests were used for either treponemal propagation or experimental ID inoculation . The Tp pertenue strain ( Gauthier ) used in these experiments was isolated in the early 1960`s in Brazzaville , Congo , from a patient’s skin lesion and provided to us by Dr . Sheila Lukehart ( University of Washington ) , who previously received it from Dr . Peter Perine ( CDC , Atlanta , GA ) . A 2012 frozen glycerol stock of the Gauthier strain containing 4x106 Tp cells/ml was inoculated into the testicles of a NZW rabbit as previously described [55] and treponemes were allowed to proliferate until the animal developed an orchitis and presence of treponemal cells within testicular tissue could be assessed by dark-field microscopy ( DFM ) from a needle aspirate . An aliquot of the glycerol stock ( 100 μl ) was saved for DNA extraction using the DNA Mini Kit ( Qiagen , Germantown , MD ) to confirm strain identity by PCR using the tprL gene ( tp1031 ) as amplification target as previously described [56] . Briefly , the amplicon generated by Tp pertenue DNA ( 209 bp ) in the tprL PCR assay differs in size from that originated by Tp pallidum and endemicum subspecies DNA ( 588 bp ) due to a deletion that encompasses part of the tprL 5’-flanking region and ORF in the pertenue subspecies . Bacteria for ID inoculations of test animals were extracted from the testes of the euthanized rabbit in sterile saline supplemented with 10% normal rabbit serum ( NRS ) . Testicular extract was collected in a sterile 15 ml tube and spun twice at 1 , 000 rpm ( 180 x g ) for 10 minutes in an Eppendorf 5810R centrifuge ( Eppendorf , Hauppauge , NY ) to remove rabbit cellular debris . Treponemes were enumerated using DFM and percentage of motile organisms was also recorded . Extract was diluted in 10% NRS-saline to obtain approximately 7 ml of treponemal suspension at the desired concentration ( 107 cells/ml ) . Test ( n = 4 ) and control rabbits ( n = 2 ) were injected ID with 100 μl of treponemal suspension ( containing 106 treponemes ) in 10 sites on their clipped backs . The skin was marked with permanent ink one inch below each injection site to facilitate location of the lesions . Following ID injection , treponemal motility was assessed again using DFM to ensure that the time elapsed between harvest and ID inoculation did not affect pathogen viability . After ID inoculation , rabbit backs were clipped daily to allow monitoring of lesion development and surgical procedures to collect lesion biopsies and aspirates . For the purpose of antimicrobial testing , ID infection is preferable to IT infection because skin lesions can be readily aspirated for DFM examination to assess the presence of Tp cells and evaluate treatment efficacy . A 0 . 5% w/v GaM cream was provided by Gallixa ( Menlo Park , CA ) along with the carrier alone ( an emulsion of water and hydrophilic petrolatum ) . Test animals were divided into three groups ( Groups 1–3 ) , each group containing two rabbits . Groups 1 and 2 received GaM twice a day every 12 hours . Group 1 rabbits began treatment at the first clinical evidence of infection by DFM analysis of lesion aspirates ( day 4 post-inoculation ) , while Group 2 rabbits received GaM when lesions had become clearly indurated ( day 14 post-inoculation ) . Administration at day 4 post-inoculation , was performed to evaluate GaM ability to prevent lesion progression , while application at day 14 post-infection aimed at evaluating GaM ability to reduce treponemal burden faster than in control animals and accelerate lesion healing . Group 3 rabbits received an equal amount of carrier only from day 4 post-inoculation . Treatment consisted in applying 250 μL of either GaM or carrier on top of each lesion , followed by gentle manual spreading to ensure uniform coverage of the whole lesion . Following application , animals were monitored for a few minutes to ensure that they would not remove the ointment , and were then taken back to their cages . Development of skin lesions at injection sites was monitored in all experimental subjects by measuring diameter of indurated lesions each day , after shaving the animals and prior to the first application of GaM or carrier . Appearance of lesions at distant sites , due to pathogen hematogenous dissemination from the primary injections sites , was also monitored . Treponemal burden within primary lesions was first assessed by performing DFM analysis of lesion needle aspirates of all lesions at day 13 post-inoculation . At day 23 post-inoculation , a second set of aspirates was obtained from Group 2 ( GaM-treated since day 14 post-inoculation ) and Group 3 ( carrier-treated control ) rabbits from all lesions ( except those that were previously biopsied ) . Approximately 100 fields per slide were examined and the treponeme number was recorded . Two lesion biopsies were obtained at day 13 post-inoculation using a 4 mm biopsy punch from each of Group 1 ( GaM-treated since day 4 post-inoculation ) and Group 3 ( control ) animals for evaluation of treponemal burden by real-time amplification of Tp mRNA ( see below ) . Two additional biopsies were obtained at day 25 and day 33 post-inoculation from each animal in Group 2 ( treated from day 14 ) and Group 3 ( control ) . Inoculation sites to be biopsied were selected randomly . A flowchart describing the experimental design is provided in Fig 1 . In all cases , biopsy samples were minced with a sterile scalpel immediately after collection and further homogenized in 400 μl of phenol-based TRIzol buffer ( Life Technologies , Santa Clara , CA ) using a disposable plastic pestle . Samples were stored at -80°C until use . Total RNA from biopsies was obtained according to the TRIzol extraction protocol . Extracted RNA was treated with DNaseI to obtain DNA-free RNA as previously described [57] . Reverse transcription into cDNA was performed with the Superscript III First-Strand Synthesis System ( Life Technologies ) using random primers according to the manufacturer's instructions . Message quantification was performed using an established relative quantification method that targets the mRNA for the treponemal 47 kDa lipoprotein ( encoded by the tp0574 gene ) and that normalizes the tp0574 signal to the message for the rabbit housekeeping gene Hypoxanthine-Guanine Phosphoribosyl Transferase ( HPRT ) [57] . Primer sequences and real-time amplification conditions for both targets , as well as details on plasmid standard preparation , were previously published [57 , 58] . Data from message quantification and treponemal counts from aspirates were analyzed with Student’s unpaired two-tailed t-test and significance set at p≤0 . 05 . All animals were bled weekly for serology ( FTA-ABS and VDRL ) , to assess seroconversion and confirm establishment of infection . Animals were euthanized 45 days post-inoculation after serological evidence demonstrated that all had become infected . Serological assays were performed on the same day for all sera collected at a given time point , to minimize test-to-test variability . The technologist performing the assays was blinded to the treatment status of the animals from which the samples were collected . Immunohistochemistry ( IHC ) . At day 13 post-inoculation a 4-mm lesion biopsy was taken randomly from each animal in Group 1 and Group 3 . Two additional biopsies from disseminated skin lesions that appeared in one of the Group 1 rabbits were also taken . All biopsies were fixed in 10 ml of 10% neutral buffered formalin ( NBF ) at room temperature for approximately 72 hours and then transferred to 70% ethanol and stored at 4°C until paraffin embedding and sectioning . For embedding , biopsies were transferred back into 4% NBF ( PanReac Applichem , Barcelona , Spain ) for 3 hrs ( 2 x 1 . 5 hr passages ) . Subsequent sample processing was performed in a Leica ASP300 instrument ( Leica Biosystems , Wetzlar , Germany ) . Samples were incubated in water for 10 min , and then transferred in 80% ethanol for 1 hr , in 96% ethanol for a total of 2 hrs ( 2 x 1 hr passages ) , and in absolute ethanol for 2 hrs ( 2 x 1 hr passages ) . Following dehydration , samples were transferred into paraffin solvent ( Histo-Clear , National Diagnostics , Atlanta , GA ) for 2 hrs ( 2 x 1 hr passages ) , followed by three passages of 1 hr each in liquid paraffin at 58°C ( Paraplast X-tra , Millipore-Sigma , St . Louis , MO ) . From these samples , 3-μm sections were cut , placed on a heath block at 65°C for a total of 20 min to allow tissue adherence to the slide , and then stored at room temperature . For IHC procedures , silane-treated slides were used to further improve tissue adherence , and tissue sections were stored at 37°C . For hematoxylin and eosin ( HE ) staining , deparaffinized and rehydrated sections were placed in hematoxylin solution for 8 min and then rinsed for 3 min with tap water . Eosin staining was carried on for 1 min prior to washing under tap water for 5 minutes . Dehydration was obtained by passage in 96% ethanol for 4 min , followed by 2 passages in 100% ethanol for 2 and 3 min , respectively , and 2 passages in Histo-Clear for a total of 3 min . Sections were mounted using an acrylic resin ( Eukitt , Orsatech , Gmbh ) , taking care to leave no bubbles during the process . Slides were left to air-dry overnight before being analyzed . For specific immunostaining , slides were first heated at 65°C for 1 hr , and then deparaffinized in EX-Prep solution ( Roche Diagnostics , Indianapolis , IN ) at 72°C . Cell conditioning was performed by applying ULTRA CC1 solution ( Roche diagnostics ) for a variable time ( 20–36 min , depending on the primary antibody ) to correct epitope alteration due to fixation of the tissue sections . Polyclonal anti-CD4 , -CD8 , and -CD20 primary antibodies ( Roche Diagnostics ) were used at 1:100 dilution , and slides were incubated for 16 min ( anti-CD4 and -CD8 ) or for 12 min ( anti-CD20 antibodies ) . To avoid evaporation , tissue sections were covered with ULTRACS liquid coverslip ( Roche Diagnostics ) following application of the primary antibody . Primary antibody was removed by washes with a Tris-based buffer solution ( Reaction buffer , pH 7 . 6; Roche Diagnostics ) . Reagents provided in the Ultra View Universal DAB ( 3 , 3’ diamino-benzidine ) Detection kit ( Roche Diagnostics ) were used according to the manufacturer`s instruction for detection of primary antibody binding . Slides were then rinsed with water and counterstained with hematoxylin for 12 minutes . Tissue sections were dehydrated using two 5-min rinses with 96% ethanol followed by two 5-min rinses in absolute ethanol , and two 10-min washes with Histo-Clean . Prior to reading , coverslips were mounted on slides using the Eukitt acrylic resin and air-dried prior to being read .
Measurements of lesion diameter as a function of time ( Fig 2 ) showed that in animals treated with GaM since day 4 post-inoculation , lesion development was significantly attenuated compared to controls ( Group 3 ) or to animals treated since day 14 post-inoculation . Most lesions form Group 1 animals failed to develop into indurated papules and enlarge , but rather remained flat , although generally erythematous . Compared to controls , only one of the Group 2 rabbits that initiated treatment at day 14 post-inoculation showed a significant decrease in lesion diameter . Assessment of treponemal burden by DFM on lesion aspirates obtained at day 13 ( Fig 3A ) and day 23 ( Fig 3B ) post-inoculation showed that compared to controls and untreated animals , treponemal burden in lesions from GaM-treated animals was significantly reduced ( p<0 . 05 ) , while carrier-treated animals and untreated did not show significant differences in number of treponemes counted . Analysis of treponemal burden performed at day 23 post-inoculation showed that overall significantly fewer treponemes could be found in lesions from rabbits that started treatment at day 14 post-infection ( Group 2 ) compared to controls . Table 1 summarizes the DFM results for each animal in each group , together with the results of FTA-ABS and VDRL tests following experimental infection . Notably , treated animals seroconverted approximately a week later than control animals , which is consistent with the reduced treponemal burden due to GaM . Treponemal burden in primary lesions was further assessed using Tp mRNA quantification normalized to the rabbit housekeeping gene HPRT . Message quantification at day 13 post-inoculation showed that no treponemal mRNA was detected from Group 1 rabbit lesions compared to controls ( Fig 4A ) . At day 23 post-infection , significantly less ( p<0 . 05 ) Tp mRNA was detected in lesions from Group 2 rabbits compared to controls , while no difference was seen between these rabbits and the control ones at day 33 post-inoculation . By the end of the experiment ( day 45 post-inoculation ) all animals had developed erythematous disseminated skin lesions . Analysis of needle aspirates from a small subset of disseminated lesions revealed the presence of treponemes by DFM ( not shown ) . Both biopsies obtained from Group 1 rabbits showed the presence of very modest inflammatory infiltrates and absence of damage to follicles ( Fig 4A and 4B ) , nearly like normal skin . Histological analysis of a disseminated lesion biopsy from one of the Group 1 rabbits ( Fig 5C and 5D ) showed a rich infiltrate of inflammatory cells , particularly eosinophils , and comparable amounts of CD4 and CD8 T-lymphocytes , plus B-lymphocytes ( CD20 cells ) and plasma cells , as well as follicular inflammation and intra-follicular abscesses ( Fig 5C and 5D ) . Analysis of a second disseminated lesion from the same animal also showed extensive follicular inflammation and an elevated number of eosinophils , although the lymphocyte component could not be evaluated due to a scarcity of cells ( not shown ) . Also , biopsies from carrier-treated animals showed a significant inflammatory infiltrate composed of eosinophils , lymphocytes ( CD4 , CD8 , and CD20 ) , and plasma cells ( CD138+ cells ) ( Fig 5E–5G ) . Biopsies from control animals also showed elevated numbers of histiocytic cells and blood vessels with a thickened endothelium ( Fig 4F and 4G ) .
The major limitation of this study was the small number of laboratory animals used , which did not allow us to make clear conclusions on the efficacy of GaM in accelerating lesion healing when administered to indurated lesions ( Group 2 animals ) . The promising results of this pilot study , however , suggest that the experiments described here should be repeated with groups of at least 8 rabbits each , according to power/sample size calculations . Furthermore , although topical GaM application was shown to have treponemicidal activity , as expected it did not prevent pathogen dissemination and establishment of the infection , even in animals that started treatment as soon as day 4 post-challenge . Our study did not address the efficacy of systemic GaM in eradicating experimental yaws . Additional studies on oral administration of GaM alone or in combination with conventional antimicrobials will need to be performed to fill this knowledge gap . Lastly , a large inoculum ( 106 cells/injection site ) was used to induce lesion development within an acceptable experimental time-frame and to obtain samples in which the treponemal burden could be quantified . Very likely , during natural human infection , significantly fewer treponemes pass the epithelial barrier to cause disease . The use of large inocula has previously been used to evaluate the effectiveness of azithromycin against Tp pallidum , and was not shown to be a confounding factor , and we have no reason to believe it could be in our studies either . Gallium has previously been shown to be effective against many microorganisms in vitro and in animal models [41 , 42] . This study is the first to extend these observations to Tp pertenue and to report the use of GaM as a topical treatment for yaws . Our results suggest that this compound could be useful as a topical anti-treponemal agent , and justifies further research into the use of GaM as both a topical and an oral agent , alone and/or in combination with other antimicrobials to assess its full potential as a novel anti-yaws compound .
|
Yaws is a neglected tropical disease affecting children in underprivileged countries , transmitted through direct skin contact with an active lesion . This infection , although rarely fatal , can lead to disfigurement and serious disability . The World Health Organization is currently conducting a yaws eradication effort that employs mass administration of azithromycin , an antibiotic against which the yaws pathogen has slowly begun to develop genetic resistance . Because this phenomenon has the potential to undermine the eradication effort , we investigated the antimicrobial activity of gallium maltolate , which has a novel mechanism of action , against the yaws pathogen . Our initial results show that topical application of gallium maltolate has significant treponemicidal activity , and suggest that this compound might find an application in the effort to eradicate yaws . Future studies will evaluate whether oral administration of gallium maltolate is as effective as the antibiotics currently approved for yaws treatment to clear systemic infection .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"antimicrobials",
"dermatology",
"urology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"biopsy",
"drugs",
"tropical",
"diseases",
"microbiology",
"rabbits",
"vertebrates",
"animals",
"mammals",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"animal",
"models",
"treponematoses",
"bacterial",
"diseases",
"skin",
"infections",
"signs",
"and",
"symptoms",
"gallium",
"antibiotics",
"sexually",
"transmitted",
"diseases",
"experimental",
"organism",
"systems",
"neglected",
"tropical",
"diseases",
"pharmacology",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"penicillin",
"animal",
"studies",
"lesions",
"chemistry",
"yaws",
"leporids",
"chemical",
"elements",
"eukaryota",
"diagnostic",
"medicine",
"genitourinary",
"infections",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"amniotes",
"organisms",
"syphilis"
] |
2019
|
Topical treatment with gallium maltolate reduces Treponema pallidum subsp. pertenue burden in primary experimental lesions in a rabbit model of yaws
|
Human cytomegalovirus ( HCMV ) is the most common cause of congenital virus infection . Congenital HCMV infection occurs in 0 . 2–1% of all births , and causes birth defects and developmental abnormalities , including sensorineural hearing loss and developmental delay . Several key studies have established the guinea pig as a tractable model for the study of congenital HCMV infection and have shown that polyclonal antibodies can be protective [1]–[3] . In this study , we demonstrate that an anti-guinea pig CMV ( GPCMV ) glycoprotein H/glycoprotein L neutralizing monoclonal antibody protects against fetal infection and loss in the guinea pig . Furthermore , we have delineated the kinetics of GPCMV congenital infection , from maternal infection ( salivary glands , seroconversion , placenta ) to fetal infection ( fetus and amniotic fluid ) . Our studies support the hypothesis that a neutralizing monoclonal antibody targeting an envelope GPCMV glycoprotein can protect the fetus from infection and may shed light on the therapeutic intervention of HCMV congenital infection in humans .
Human cytomegalovirus ( HCMV ) , a member of the herpesvirus family , is widely distributed in the human population and can cause severe disease in immunocompromised patients and upon infection of the fetus . A therapeutic for HCMV infection is a major public health priority for women with primary HCMV infections during pregnancy . Congenitally-infected infants have a high incidence of neurodevelopmental sequelae , including mental retardation and sensorineural deafness [4]–[7] . Several lines of evidence suggest that neutralizing antibodies can protect the fetus from HCMV infection and disease . First , preconception maternal antibodies to HCMV significantly reduce the severity and risk of congenital HCMV infection in future pregnancies , although the frequency of sensorineural deafness is the same [6] , [8] . Second , early appearance of maternal antibodies against the HCMV glycoprotein entry complex , gH/gL/UL128/UL130/UL131 , correlates with a lack of vertical transmission , suggesting that antibodies against this complex can effectively neutralize the virus in vivo [9] , [10] . And third , a small , open-label study showed that hyperimmuneglobulin could protect the fetus from HCMV infection and disease [11] . However , hyperimmuneglobulin is not an optimal therapy because of lot-to-lot variability , the possibility of inadvertent transmission of infection , the volumes that must be administered , and difficulties in maintaining an adequate supply . A monoclonal antibody with therapeutic efficacy would overcome all these problems , but there is no proof of concept that antibody against a single CMV epitope could confer protection against fetal infection . Therefore , we set out to test this hypothesis . The guinea pig has been a useful model for the study of maternal-fetal transmission because the placental anatomy is similar to that of humans [12]–[15] . However HCMV does not infect guinea pigs , thereby necessitating the use of guinea pig CMV ( GPCMV ) . GPCMV has been demonstrated to cross the placenta and cause fetal infection . Several studies have delineated a role for antibodies in the prevention of GPCMV infection and disease in the guinea pig [1]–[3] In early studies , Bia et al . demonstrated that preconception infection protected against in utero GPCMV transmission , most likely due to the generation of neutralizing antibodies in the mother [16] , [17] . More recently , Bourne et al . , determined that preconception immunization of pregnant guinea pigs with GPCMV glycoproteins protected the fetus from death and infection [2] . In addition , two studies determined that passive immunization of anti-GPCMV antibodies could be protective in the congenital setting [1] , [3] . In this study , we determine whether a monoclonal antibody can confer protection in the congenital infection setting . To this end , we generated neutralizing monoclonal antibodies against GPCMV . We show that a neutralizing monoclonal antibody against the GPCMV gH/gL glycoprotein entry complex , when administered prophylactically to mothers with primary infection , protects against fetal infection and death . In addition , we characterize the kinetics of congenital infection , from maternal infection ( seroconversion , salivary glands and placental infection ) to fetal infection ( fetus and amniotic fluid ) and find that congenital infection is rapid following infection of the mother . These studies support the hypothesis that a neutralizing antibody response that targets a single epitope on a glycoprotein entry complex can be protective in the context of congenital infection .
To understand the timing of seroconversion during pregnancy , GPCMV-free pregnant guinea pigs were inoculated with 4×103 PFUs of pathogenic salivary gland-passaged GPCMV ( in vivo passage 8 , IVP8 ) at day 21 of gestation ( beginning of the second trimester ) . 4×103 PFUs was empirically determined to result in robust fetal infection with minimal dam mortality ( data not shown ) . Antibodies against GPCMV soluble gB or gH/gL protein could be detected by day 7 , indicating primary infection , and continued to rise throughout the study ( Figure 1 ) . Reactivity against gB protein in serum samples tested at 1∶100 dilution appeared to plateau at day 10 . However , additional analysis of samples diluted to 1∶2700 revealed that reactivity against gB protein continued to increase beyond day 10 ( Figure 1A ) . Though minimal , serum reactivity against soluble gH/gL antigen gradually increased over the course of the study as well ( Figure 1B ) . To better understand the kinetics of maternal , placental , and fetal infection , we performed a time course study of infection ( Figure S1 ) . 43 pregnant guinea pigs were inoculated with 4×103 PFU of GPCMV IVP8 at day 21 gestation and sacrificed at 1 , 3 , 7 , 11 , 15 , and 21 days post-infection . Guinea pigs were infected at day 21 of gestation to allow for enumeration of viable fetuses by ultrasound ( Video S1; Methods for details ) . Infection of dams and fetal tissues were determined by PCR analysis . Due to the limitations on the number of GPCMV-free , timed-pregnant animals that could be obtained at a given time , this study was performed as three separate studies with overlapping time points . Given that the kinetics of fetal infection ( i . e . slopes of the curves ) was similar among the three studies ( Figure 2 ) , the data from the three studies was combined . Dam mortality ( 1/5 ) occurred at d11 post-infection , and increased to 38% mortality overall on d21 ( Table 1 ) . Neutralizing titer in dams increased over the time course , with minimal titers at days 7 and 11 and robust titer at day 21 ( Table 2 ) . Fetal loss was observed as early as d1 ( 2/19 , 10% ) , and preceded dam mortality , and increased to 74% loss overall by d21 ( Table 1 ) . Fetal loss early in the study is likely not attributable to viral infection due to the fact that a similar rate of fetal loss was found in uninfected controls ( Table 1 ) . As early as d7 , viral genomes ( as measured by qPCR ) were detected in the maternal salivary glands in 100% ( 5/5 ) of the animals , with an average copy number of 1×102 copies/salivary gland . Also at d7 , viral genomes were detected in 50% ( 20/40 ) of placentas from infected dams which increased to 90% ( 18/20 ) by d11 ( Table 2 ) . Consistent with the literature [18] , viral genome number peaked in maternal blood at d11 with 1×104 copies/mL , followed by a rapid reduction by d15 ( data not shown ) . Peak viral load ( d11 ) was coincident with the first maternal death . Using qPCR , viral genomes were not detectable in the fetus or the amniotic fluid . However , when a more sensitive nested PCR assay was employed , viral genomes could be detected in the fetus as early as d7 , with 31% ( 12/39 ) of the fetuses infected ( Table 2; Figure 3 ) . Moreover , when the nested PCR assay was applied to the maternal salivary glands and the placenta , virus could be detected robustly at d3 post-infection ( Table 2; Figure 3 ) . However we can not formally rule out that detection of virus in the placenta early in infection is due to contamination from maternal blood . When the incidence of placental and fetal infection from individual mothers was plotted over time , the following pattern emerges: placental infection occurred without detectable fetal infection ( e . g . glyphs plotted below the diagonal line ) but fetal infection did not occur without detectable placental infection ( e . g . lack of glyphs plotted above the diagonal line ) ( Figure 3 ) . This pattern suggests a temporal relationship between placental and fetal infection with placental infection occurring prior to fetal infection ( Figure 3 ) . In addition , using the nested PCR assay , viral genomes could be detected in the amniotic fluid by day 11 , with 81% ( 13/16 ) infected ( Table 2 ) . Given that the first detectable amniotic fluid infection occurred following fetal infection , these results suggest a sequential order of infection , beginning with maternal salivary gland infection and rapidly spreading to the placenta , the fetus , and then the amniotic fluid . The goal of this study was to evaluate the ability of a neutralizing monoclonal antibody to protect dams and their offspring from infection and loss . GPCMV encodes homologs of the HCMV glycoproteins that mediate viral entry: gB , gH/gL , gO , in addition to UL128 , UL130 , and UL131A ( referred to in GPCMV as GP129 , GP131 , and GP133 ) [19]–[25] . In order to develop monoclonal antibodies that neutralize GPCMV , mice were immunized with GPCMV virions or baculovirus-expressed recombinant glycoprotein complexes . The resulting clones were screened for GPCMV neutralizing activity on primary guinea pig fibroblast and endothelial cells . Using this approach , we screened over 5000 hybridoma clones and isolated 6 hybridoma clones that neutralized GPCMV with EC50s ( µg/mL ) ranging from 0 . 01 to 2 . 1 on endothelial cells and 0 . 07 to 4 . 1 on fibroblast cells ( Table 3 ) . To determine the antigen specificities of each antibody , binding studies were performed on the following soluble proteins: gB , gH/gL , gH/gL/gO complex , and gH/gL/GP129-133 complex . Binding analysis with soluble glycoprotein complexes revealed that three antibodies recognize the gH/gL heterodimer ( 1597 , 1968 , 394 ) and one clone ( 1282 ) recognizes gH/gL/gO but not gH/gL alone , suggesting that the antibody specifically recognizes gO . The remaining two antibodies ( 1778 , 1593 ) did not react with any of these complexes ( Table 3 ) . Further confirmation of these results with antibodies 1597 , 1968 , and 394 was obtained using FACS analysis on cell surface-expressed gH/gL complex ( Table 3 ) . Competition FACS experiments with the anti-gH/gL monoclonal antibodies revealed that 1597 and 1968 compete with each other ( data not shown ) . Extensive characterization of the most potent antibodies , 1778 and 1593 , did not reveal their epitopes , and as such , these antibodies were not pursued further . Neutralizing antibodies were not recovered that recognized GPCMV gB or GP129/GP131/GP133 . We moved forward with anti-gH/gL clone 1968 , due to the fact that it is the most potent antibody with known target specificity . In order for a monoclonal antibody to protect the fetus from infection and/or disease , it needs to be present in the maternal serum . However , a mouse IgG may have poor pharmacokinetics ( PK ) in the guinea pig for two reasons: 1 ) mouse IgG does not bind significantly to human FcRn , and by extension may not bind to guinea pig FcRn [26] , and 2 ) the guinea pig may mount an immune response against the foreign mouse antibody , thus rapidly clearing it from circulation . To improve PK , chimeras between the mouse F ( ab′ ) 2 and the guinea pig IgG2 constant region were generated ( Figure 4A ) . An irrelevant isotype control directed against HIV gp120 antigen was constructed in parallel ( Figure 4A ) . In a neutralization assay on fibroblast and endothelial cells , the 1968 mouse-guinea pig chimeric antibody ( henceforth referred to as 1968/GPFc ) retained anti-GPCMV potency similar to the fully-murine 1968 antibody ( Figure 4B ) . 1968/GPFc was also able to neutralize the pathogenic GPCMV stock ( IVP8 ) on fibroblasts with potency similar to the tissue culture-adapted GPCMV strain ( strain 22122; Figure S2 ) . To evaluate the ability of the 1968/GPFc antibody to bind guinea pig FcRn , binding assays were performed . To this end , the guinea pig genomic sequences encoding for the extracellular domain of FcRn and β2-microglobulin were cloned and transiently co-expressed in HEK293 cells ( Figure S3 ) . Both the 1968/GPFc and the anti-gp120 chimera were evaluated for their ability to bind the FcRn-β2-microglobulin soluble complex in biosensor assays . Both antibodies bound to FcRn in equal affinity to guinea pig purified IgG from serum , whereas the mouse antibody 1968 failed to bind ( Figure 4C ) . PK of the 1968/GPFc antibody in guinea pigs was evaluated following a single antibody dose at 10 mg/kg in infected and uninfected pregnant dams ( Figure S4 ) . The 1968/GPFc antibody was administered at 1-day post-infection . Blood samples were drawn from each guinea pig before the start of the study and at 0 . 5 h , 8 h , d1 , d7 , d14 , and d21 post-infection . The 1968/GPFc antibody was detectable in the blood with a half-life of approximately 8 . 46 days in uninfected guinea pigs ( Figure S4 ) . This half-life is considered within normal range for an IgG in guinea pig [27] . Neutralization potency was similar for infected and uninfected animals at day 1 ( uninfected , 1∶477; infected , 1∶355 ) and day 3 uninfected 1∶841; infected 1∶622 ) . We observed an increase of anti-gH/gL antibody concentration after day 15 post-antibody administration only in infected animals . But given that this increase only occurred in infected animals , it is likely due to the presence of maternal anti-gH/gL antibodies in response to the infection . This rise in endogenous anti-gH/gL antibody titer is consistent with our observations from the seroconversion study ( Figure 1B ) . Given that the 1968/GPFc antibody is potently neutralizing against GPCMV and can bind to guinea pig FcRn , its ability to protect against maternal and fetal death as well as fetal infection was evaluated . To this end , 7 pregnant dams received the 1968/GPFc antibody at 8 mg/kg starting at one day prior to viralinoculation , followed by twice per week injections for a total of 6 doses . The 1968/GPFc antibody was administered as a prophylactic , due to the narrow therapeutic window revealed by the kinetics study . This dose of antibody was estimated to provide a serum concentration >10X in vitro neutralization EC90 at Ctrough ( i . e . trough plasma concentration measured at the end of a dosing interval at steady state ) . In parallel , 8 pregnant dams were administered the anti-gp120 chimera at the same dose for comparison . All 15 pregnant guinea pigs were inoculated with 4×103 PFU of GPCMV IVP8 at day 21 of gestation and sacrificed at day 21 post-infection ( Figure 5 ) . Between days 13 and 14 , there was mortality in half of the pregnant dams ( 4/8 ) in the anti-gp120 group ( Table 4 ) . A higher percentage of maternal loss was observed in this study than from in-house historical studies in which the average maternal loss at day 14 was 32% ( 10/31 dams ) . In contrast , only 1 out of 7 guinea pigs in the 1968/GPFc group died ( Table 4 ) . Although these findings did not reach statistical significance , the reduced death in the 1968/GPFc group is suggestive of dam protection . Of the surviving dams at day 21 post-infection , the average neutralizing viral titer was similar in both groups ( 1∶840 for 1968/GPFc-dosed animals and 1∶540 for anti-gp120-dosed animals ) . In addition to mother mortality , fetal loss and infection were also measured . 100% fetal loss was observed in the anti-gp120 group , whereas only 65% of the fetuses were lost at the time of sacrifice in the 1968/GPFc group ( Table 4 ) . Since all of the fetuses were lost in the anti-gp120 control group , the infection rate of fetuses from the 1968/GPFc group was compared to those from in-house historical controls . Two in-house studies of congenital infection were averaged to provide a historical reference ( Table 4 ) . The 1968/GPFc group showed a significant reduction in fetal infection , with 15% of the fetuses infected as compared with 60% of historical controls ( p = 0 . 03 ) ( Table 4 ) . Although protection was observed , it is noted that the fetuses from the 1968/GPFc group were recovered from two dams , providing a limited data set from which to derive a p-value . The 1968/GPFc antibody did not appear to block viral placental entry , with 100% of the placentas infected in the 1968/GPFc group . These results suggest that the presence of 1968/GPFc neutralizing antibodies reduce the rate of GPCMV fetal infection and loss , and in addition , may improve dam survival .
A therapeutic for HCMV disease is a major public health priority given the disability in newborn infants caused by congenital infection . Both in humans and in guinea pigs , passive administration of anti-CMV polyclonal antibodies has been shown to be protective to the fetus [1] , [11] . A neutralizing monoclonal antibody has not been evaluated in clinical trials for this indication nor has been tested in an animal model of congenital infection . In this study , we evaluated the ability of an anti-gH/gL neutralizing monoclonal antibody to protect against fetal loss and infection in the guinea pig model of congenital CMV . Several guinea pig studies have delineated a role for antibodies in protecting the fetus from GPCMV infection and disease [1]–[3] . In early studies , Bia et al . demonstrated that preconception infection protected against in utero GPCMV transmission , most likely due to the generation of neutralizing antibodies in the mother [17] , [28] . More recently , studies determined that passive immunization of polyclonal anti-GPCMV antibodies ( or anti-gB polyclonal antibodies ) were protective in the congenital setting [1] , [3] . Our studies build on these pioneering studies and demonstrate that a neutralizing monoclonal antibody can show efficacy in the congenital setting . These studies support the hypothesis that a neutralizing antibody response that targets a single epitope can be protective against fetal infection . By extension , a neutralizing monoclonal may be efficacious in the congenital setting in humans as well . In this study , we delineated the kinetics of seroconversion andcongenital infection . We found that seroconversion and maternal-fetal transmission of GPCMV occurred at approximately in the same time frame ( <7 days ) . Our results are consistent with the time course of transmission established by Griffiths et al . , ( 1980 ) , with the peak of fetal infection occurring at 11–15 days post-infection [29] . However , our study increases the understanding of the kinetics of primary infection via greater resolution of time points and the utilization of molecular techniques . The rapid viral spread to the placenta necessitated a prophylactic rather than therapeutic study . Altering the infection route or significantly reducing the inoculum titer to reduce the rate of viral spread unfortunately did not result in robust fetal infection ( data not shown ) . In fact , viral kinetics may in part explain the marked reduction of protection observed by Bratcher et al . , ( 1995 ) , when directly comparing the therapeutic versus prophylactic administration of hyperimmuneglobulin in guinea pigs [1] . We isolated and characterized six potently neutralizing monoclonal antibodies against GPCMV: three against the gH/gL protein , one against gO protein , and two with unknown targets . One of the anti-gH/gL antibodies , 394 , was isolated from mice immunized with gH/gL protein rather than whole virus . It is interesting to note that antibody 394 has significantly less neutralization potency than the other anti-gH/gL antibodies resulting from whole virus immunization . This suggests that viral soluble complexes may not serve as the most effective immunogens for the generation of neutralizing antibodies . Surprisingly , we did not recover any neutralizing antibodies against gB or the GP129/GP131/GP133 proteins . Antibodies against the HCMV homologs of these proteins have been shown to be highly neutralizing [30] , [31] . However , one formal possibility is that our most potent antibodies ( 1778 , 1593 ) may recognize epitopes on gB or the GP129/GP131/GP133 complex that are not identified by soluble or cell-associated complexes . Our inability to recover neutralizing antibodies against this complex may reflect that GPCMV gH/gL/UL129/UL131/UL133 complex is not essential for viral entry in fibroblast and epithelial cells ( we conducted our screens on these cell types ) [23] . Here , we have shown that a highly neutralizing monoclonal antibody against GPCMV gH/gL reduces fetal infection rate and death following maternal GPCMV challenge . Despite the protection observed , we were surprised by the inability of the anti-gH/gL monoclonal antibody , 1968/GPFc , to protect against placental infection . In humans , infection of the fetus most likely occurs by transcytosis across the placenta , in which the virus takes advantage of low avidity antibodies for transport [32] . High avidity , neutralizing antibodies have been shown to intercept this pathway , and ultimately may prevent fetal infection [32] . Mothers with preconception immunity to HCMV have babies with a much lower incidence of infection and disease , suggesting that maternal antibodies can be protective in this setting . In contrast , placental infection in the guinea pig is rapid and direct following subcutaneous inoculation . In a non-laboratory setting , spread between animals may result from a lower titer of inoculum via mucosal membranes and may ultimately lead to a slower progression of the infection , possibly allowing better protection by antibodies . A second possible explanation is that full protection may require administration of an antibody cocktail . Consistent with this possibility , Chatterjee et al , found that administration of high titer anti-gB polyclonal serum , which presumably binds multiple neutralizing epitopes , significantly reduced placental infection and prevented fetal infection [3] . However , there are limitations in drawing such comparisons due to differences in study methodology . In humans , humoral responses to natural infection set the bar for vaccines or immunotherapeutics as natural immunity is known to prevent and reduce disease . Along these lines , we found that the serum neutralizing activity of guinea pigs administered with 1968/GPFc on d21 of gestation was approximately equal to that of infected animals treated with control antibody on d21 post-infection ( i . e . d42 of gestation ) . In addition , we found that this serum neutralizing activity was similar to that from GPCMV positive guinea pigs procured from our vendor ( data not shown ) . Indeed , the concentration of anti-gH/gL antibody in guinea pigs administered with 1968/GPFc is similar to infected control guinea pigs on d21 ( 200 ug/ml for guinea pigs administered 1968/GPFc vs 120 ug/ml for guinea pigs administered control antibody ) . Despite these similarities at d21 , the key difference between animals developing their own immunity to GPCMV versus the administration of 1968/GPFc is timing: animals administered 1968/GPFc have a neutralizing serum titer significantly higher than that of naïve animals from the onset of infection , thereby providing immediate protection to the developing fetus . Along these lines , guinea pigs that are seropositive prior to conception , and thus have neutralizing titers throughout gestation , also result in protection of the fetus [29] . Given that passive antibody administration of an anti-gH/gL monoclonal antibody can provide levels of neutralizing titers similar to that of seropositive animals , these results may have important implications therapeutic development . An anti-HCMV neutralizing monoclonal antibody has not been evaluated in clinical trials for the prevention of congenital HCMV . However , MSL-109 , a neutralizing monoclonal against HCMV gH , has been evaluated in the prevention of HCMV infection following allogenic hematopoietic stem cell transplantation and as adjuvant therapy for HCMV retinitis in HIV-infected individuals without apparent benefit [33] , [34] . Recently , a nongenetic mechanism of generation of viral resistance was demonstrated in vitroand was proposed to explain the apparent clinical failure of MSL-109 [35] . However , analysis of a subset of the transplantation patients who were at high risk for primary HCMV infection ( donor positive/recipient negative patients , D+/R− ) actually demonstrated that MSL-109 could confer protection [33] . Since HCMV congenital infection can also result from primary HCMV infection in the mother , this suggests that monoclonal antibody therapy might be useful in this setting . Given that our study did not directly compare antibodies neutralizing different glycoprotein epitopes , we are not claiming that an anti-gH/gL antibody is the optimal therapy for congenital HCMV . However , a recent vaccine study strongly suggests that the pentameric gH complex is the primary target for neutralizing antibodies [36] . Our study suggests that a therapeutic that targets a single neutralizing epitope on HCMV ( or alternatively , a cocktail of monoclonal antibodies ) may have clinical benefit . Obviously , it will take a human trial to determine if a monoclonal antibody is indeed protective , but the results from this study are consistent with this hypothesis .
All animal work has been conducted on an approved protocol , reviewed and approved by Genentech's Institutional Animal Care and Use Committee ( IACUC studies: 11-2904 , 10-1827 , 10-1006A , 10-0499A ) . Genentech , Inc . is registered with the USDA and its protocols adhere to the USDA regulation of the Animal Welfare Act and Animal Welfare Regulations . Genentech , Inc . is OLAW ( Office of Laboratory Animal Welfare ) assured and protocols adhere to the Public Health Service ( PHS ) Policy on the Humane Care and Use of Laboratory Animals . Timed-pregnant Hartley guinea pigs were obtained at 15 days of gestation from Elm Hill Breeding labs ( Tynsboro and Chelmsford facilities in MA ) and housed under conditions approved by the American Association of Accreditations of Laboratory Animal Care Committee . As guinea pigs do not have a genital plug , Elm Hill Labs determines the gestation stage by the timing of the previous birth: guinea pigs typically breed 24 hours post-birth if a male is placed in the cage with the female . In addition , Elm Hill labs palpates the females to confirm pregnancy . Prior to their use , all animals were determined to be GPCMV-free by neutralization assay . In addition , we confirmed pregnancy in-house via ultrasound by which we only used guinea pigs that were: 1 ) pregnant , 2 ) had fetuses of a similar stage and size ( this can be measured using ultrasound ) , and 3 ) had viable fetuses . Of note , most pregnant females had fetuses of similar size appropriate of the expected gestational stage . Ultrasound imaging was performed using an Acuson Sequoia C512 ultrasound imaging system with a 15L8-S probe ( Siemens Medical Solutions , Malvern , PA , USA ) or Vevo2100 imaging system with a 33 MHz probe ( Visualsonics , Toronto , CN ) . Male Hartley guinea pigs obtained from the same facilities were used for in vivo passaging of the virus . HEK293T cells were obtained from ATCC ( CRL-11268 ) and cultured in DMEM+10% FBS . CHO cells were obtained from ATCC and cultured in F-12K media+10% FBS . Guinea pig embryonic fibroblasts ( gefs ) and primary endothelial cells were generated in-house according to Auerbach et al . ( 2013 ) [23] GPCMV ( strain 22122 , American Type Culture Collection ) was propagated on gefs . Viral stocks were titered at 24 hours post-infection ( hpi ) using immunofluorescence microscopy with an anti-GPCMV gB monoclonal antibody ( 29-29; gift of Bill Britt , University of Alabama ) [23] , [24] . Viral stocks had infectivity titers of 105–106 PFU/ml on fibroblast and endothelial cells . This viral stock was used to prepare the salivary gland-derived GPCMV stock of higher virulence by 8 sequential passages in vivo in Hartley male guinea pigs as previously described [37] . We refer to this pathogenic stock as IVP8 . Briefly , guinea pigs were infected subcutaneously with approximately 105 PFU . After 21 days of infection , the salivary gland was removed , homogenized , sonicated , clarified , titered and used for re-infection . After 8 sequential passages , salivary gland extract was stored in frozen aliquots at −80°C at 1∶1 in 0 . 2M sucrose phosphate buffer at 105 PFU/ml . A single pool of extract was used throughout this study . 15 pregnant guinea pigs were inoculated by subcutaneous injection of 4×103 PFU of IVP8 stock at 21 days gestation . Blood samples ( approximately 400 µl each ) were collected for ELISA analysis via the orbital route before the start of the study ( pre-immune negative control samples ) and at d3 , d7 , d10 , d14 , d17 , and d21 post-infection . 15 pregnant guinea pigs were divided into two groups to alternate bleeding: Group 1 ( n = 7 ) was bled on d3 , d10 , d17 , and Group 2 ( n = 8 ) on d7 , d14 . Guinea pigs from both groups were bled and sacrificed at d21 post-infection . Four pregnant guinea pigs died before the end of the study between d12-17 resulting in 27% dam mortality ( similar mortality to our in-house historical studies ) . The pre-immune uninfected pregnant guinea pig serum was used for negative controls to determine background levels . Serum samples were diluted in PBS+0 . 5% BSA+0 . 25% CHAPS+5 mM EDTA+0 . 35M NaCl+10 ppm Proclin+0 . 05% Tween 20 at pH 7 . 4 and analyzed by ELISA using soluble GPCMV gB or gH/gL protein ( see ELISA section for details ) . Serum interference was evaluated and donkey anti-guinea pig IgG ( H&L ) provided the best signal . Samples were diluted starting at 1/100 and then serially diluted 3-fold for a total of 8 points . The minimum dilution was found to be 1/2700 for the anti-gB IgG response and 1/100 for the anti-gH/gL IgG response . Relative OD readings were plotted in lieu of absolute antibody concentrations because of the challenges in obtaining polyclonal anti-gB and anti-gH/gL purified standards . Pre-immune samples ( e . g . serum from uninfected pregnant guinea pigs ) and day 12 positive serum were included as controls in each assay . This allowed for normalization across all ELISA plates . 43 pregnant guinea pigs were inoculated by subcutaneous injection with 4×103 PFU of pathogenic stock IVP8 at day 21 gestation and sacrificed at 1 , 3 , 7 , 11 , 15 , and 21 days post-infection . Guinea pigs were infected at day 21 of gestation to allow for enumeration of viable fetuses by ultrasound . Infection of dams was verified post-sacrifice by determination of viral copy number from maternal salivary gland homogenates . Due to the limitations on the number of GPCMV-free , timed-pregnant animals that could be obtained at a given time , this study was performed as three separate studies with overlapping time points . Despite experiment-to-experiment variation in mother survival and fetal loss in groups sacrificed on different days , the kinetics of fetal infection ( i . e . slope of the curves ) was similar among the three experiments , allowing for studies to be combined ( Figure 2 ) . In parallel , 8 dams were mock infected ( with DMEM media ) and sacrificed at 21 days post-infection . All anti-GPCMV antibodies from hybridoma screens were evaluated by neutralization assays based on those with HCMV as described by Abai et al . ( 2007 ) . Briefly , antibody was serially diluted in complete media and mixed with virus diluted in complete media such that the final virus concentration resulted in approximately one infectious virus per cell ( Multiplicity of Infection ( MOI ) = 1 ) when mixed with media . Antibody and virus were mixed and incubated at 37°C for 1 hour prior to incubation for 24 hours on a confluent monolayer of cells . 24 hours post-infection , cells were fixed with 100% ethanol and blocked in PBS and 2% bovine serum albumin ( BSA ) and then stained with monoclonal anti-guinea pig gB antibody ( 29-29; gift of Bill Britt , University of Alabama ) [24] . Cells were washed with PBS and incubated with the appropriate AlexaFluor 488 and Hoechst ( Invitrogen ) stains . Cells were imaged and counted using the ImageXpress Micro and MetaXpress software ( Molecular Devices; Sunnyvale , CA ) . All guinea pig sera ( e . g . serostatus screening; kinetic analysis of infection; protection study ) and tissue homogenate ( e . g . neutralization of IVP8 by 1968/GPFc ) were evaluated for presence of neutralizing antibodies by an in-well-lysis-TaqMan qPCR assay . In-well-lysis-TaqMan qPCR assay was used as an alternate to the immunofluorescence assay due to the high background when working with serum and tissue . Serum was serially-diluted in 5-fold dilutions in DMEM media , and mixed with virus ( MOI = 1 ) . The mixture was incubated for an hour at 37°C prior to incubation for 48 hours on a confluent monolayer of guinea pig primary cells . Cells were lysed using the Cells-to-CT kit ( Ambion , Austin , TX ) and the relative percentage of infection was quantified in reference to negative controls ( e . g . , no serum , uninfected guinea pig serum ) . A multiplex qPCR using TaqMan probes ( Applied Biosystems , Carlsbad , CA ) targeting the GPCMV gene GP83 and the endogenous control , guinea pig β-actin , allowed for analysis using the comparative ΔΔCT method [38] . qPCR reactions were performed in an ABI 7500 real-time PCR system and then analyzed using the ABI analysis software ( Applied Biosystems ) . See the qPCR assay section below for primer sequence and PCR conditions . All assays were carried out in duplicate , and the results are expressed as the normalized percent of infection . Neutralization data was analyzed with Prism 5 . 0 ( GraphPad Software; La Jolla , CA ) using non-linear regression ( curve fit; 4-parameter ) . Ten Balb/c mice ( Charles River Laboratories International , Inc . , MA , USA ) were immunized intraperitoneally with 106 PFU/mice/injection of GPCMV whole virus ( strain 22122 ) in an adjuvant containing metabolizable squalene ( 4% v/v ) , Tween 80 ( 0 . 2% v/v ) , trehalose 6 , 6-dimycolate ( 0 . 05% w/v ) and monophosphoryl lipid A ( 0 . 05% w/v; all components obtained from Sigma Aldrich , USA ) . In addition , in order to generate antibodies against the gH/gL and gH/gL/GP129/GP131/GP133 complexes , we immunized a subset of mice with soluble gH/gL or gH/gL/GP129/GP131/GP133 proteins at 2 µg of protein/mouse using the same adjuvant that was used for the whole virus immunizations [23] . The mice were boosted with the virus or soluble protein , with adjuvant , twice per week . Following 10 injections , serum samples were evaluated for viral neutralizing activity in vitro using primary guinea pig endothelial cells . B cells from spleens harvested from three mice demonstrating neutralizing serum activity were then fused with mouse myeloma cells ( X63 . Ag8 . 653; American Type Culture Collection , Manassas , VA , USA ) by electrofusion ( Hybrimmune , ECM 2001; Harvard Apparatus , Inc . , Holliston , MA , USA ) . After 10–14 days the supernatants were harvested and screened for IgG production by a direct ELISA . All ELISA positives ( i . e . cells producing IgG ) were re-screened to evaluate viral neutralizing activity . Clones demonstrating the desired neutralizing activity were then subcloned by limiting dilution ( single cell/well ) and retested as described above . The final clones were cultured in INTEGRA CELLine 1000 bioreactors ( INTEGRA Biosciences AG , Zizers , Switzerland ) . The supernatants were then purified by affinity chromatography ( MabSelect SuRe; GE Healthcare , Piscataway , NJ , USA ) , sterile-filtered ( 0 . 2 µm ) , and stored at 4°C in PBS . All monoclonal antibodies were characterized by ELISA and FACS ( see Methods sections below ) . The production and characterization of recombinant GPCMV gB , gH , gL , GP129 , GP131 , and GP133 proteins have been described in Auerbach et al . ( 2013 ) [23] . Briefly , each gene was cloned into pAcGP67 for expression in the baculovirus system with a C-terminal His tag . Native signal sequences were removed and replaced with the insect signal sequence . gB , and gH transmembrane domains were eliminated to maximize secretion into the media . Plasmids were transfected into Spodoptera frugiperda ( Sf9 ) and Trichloplusia ni ( Tni ) cells ( Expression Systems LLC ) , passaged 3 times and stored in 2% HIFBS . For protein expression , cells were infected , incubated at 37°C and supernatant was harvested at 72 hrs post infection . Protein was then purified over nickel resin and analyzed by SDS-PAGE and Western blot . 96- or 384-well Nunc Maxisorp plates were coated with the following baculovirus-generated proteins: gB at 2 µg/ml , gH/gL at 0 . 5 µg/mL , gH/gL/gO at 0 . 1 µg/mL , or gH/gL/GP129/GP131/GP133 0 . 1 µg/mL in 0 . 05M Sodium Carbonate Buffer , pH 9 . 6 . Plates were washed three times with PBS/0 . 05% Tween 20 and blocked with 50 µL of PBS/0 . 5%BSA/10 ppm Proclin . Guinea pig serum samples or anti-GPCMV monoclonal antibodies were allowed to bind to viral antigens for two hours and the plates were rinsed six times with wash buffer . Goat anti-guinea pig IgG ( H&L ) conjugated to horseradish peroxidase ( HRP ) ( Jackson ImmunoResearch ) at 40 ng/ml was added for one hour to detect the GPCMV-specific IgG antibodies . The plates were washed 6 times prior to addition of and TMB substrate ( Moss Inc; Pasadena , MD ) . The reaction was stopped after 10 minutes with equal volume 1M phosphoric acid and absorbance was read at 450 nm and referenced at 620 nm . Plasmids containing gH+gL glycoproteins from GPCMV were constructed such that each protein was expressed in equal stoichiometry by separating each gene with a “self-cleaving 2A peptide” [39] . The plasmid contains GPCMV gH+gL+eGFP ( cloned from cDNA ) . Plasmid was transfected into human embryonic kidney ( HEK ) -293T cells ( ATCC ) using Lipofectamine 2000 ( Invitrogen; Carlsbad , CA ) to express GPCMV gH/gL complex at the surface . After 2 days , cells were dissociated and stained . Fluorescence of individual cells was measured using FACSCalibur ( BD Biosciences ) and analyzed using FlowJo software ( Tree Star ) . At the time of this work , only the protein sequences of the guinea pig IgG Fc regions were published . The protein sequences of the constant region of guinea pig IgG1 , human IgG1 , murine IgG2a and rabbit IgG1 were aligned and used to design degenerate PCR primers to regions of homology . Once a fragment of the constant region was cloned , 5′ and 3′ RACE PCR was used to determine the nucleotide sequence of the entire constant region . Two guinea pig antibody heavy chain isotypes ( IgG1 and IgG2 ) were identified . Due to its sequence similarity to human IgG1 , guinea pig IgG2 was used . The nucleotide sequence for IgG2 was submitted to the GenBank database and assigned accession numbers KF491482 ( heavy chain ) and KF491483 ( light chain ) . The variable heavy and light chain domains of hybridoma cell line 1968 were cloned directly from cells using a 5′RACE protocol . The PCR amplified products were subcloned into TOPO-TA ( Invitrogen , Invitrogen , Carlsbad , CA ) transformed into bacteria and plated on agar . Individual colonies were propagated to obtain plasmid DNA from which the DNA sequences of the subcloned heavy and light chain could be determined . Based on the sequencing information , nested PCR primers were designed to allow restriction digest free cloning of the heavy and light chain domains into mammalian expression vectors encoding the guinea pig IgG2 heavy constant and kappa constant region . The restriction digest free cloning ensured that the entire murine variable heavy and light chain were inserted without sequence modification and seamlessly joined to the guinea pig constant domains . The 1968 mouse-guinea pig chimeric antibody is referred to as 1968/GPFc . A negative control murine-guinea pig IgG2 chimeric plasmid was assembled in a similar fashion . Here , restriction free cloning was initiated from earlier cloned heavy and light chain expression plasmids of a murine antibody directed against gp120 of HIV-1 . The DNA sequence of the guinea pig β2-microglobulin was available from Genbank ( accession number AF148875 . 1 ) . PCR primers were designed for the amplification of guinea pig β2-microglobulin from a commercially available liver cDNA library . The β2-microglobulin gene was amplified by PCR , cloned into a mammalian expression vector , and sequenced . The sequence of the extracellular domain of guinea pig FcRn was identified by PCR with degenerate oligonucleotides designed from regions of homology based on an alignment of rabbit , human , cynomolgus monkey , bovine , sheep , rat and mouse FcRn . The final product from 5′ and 3′ RACE PCR was cloned into a mammalian expression vector and co-transfected into CHO cells with the β2-microglobulin gene ( gene sequence obtained from Genbank ) . Cells were allowed to express the FcRn complex for 5 days . Purified guinea pig FcRn complex was acquired by filtering the cell supernatants , then affinity purifying on human IgG columns at pH 6 . Mass spectrometry of a deglycosylated and reduced sample showed the expected molecular weights of the two chains . The nucleotide sequence for guinea pig FcRn has been submitted to the GenBank database and assigned accession number KF491481 . The binding of antibodies to guinea pig FcRn complex was evaluated using an Octet Red QK instrument: a real-time , label-free platform that evaluates protein-protein interaction . Briefly , the monoclonal antibodies were diluted to 30 µg/mL in 2- ( N-morpholino ) ethanesulfonic acid ( MES , pH 5 ) buffer and were coupled onto amine reactive biosensors ( ForteBio ) for 15 minutes . The excess reactive sites on the sensors were blocked for 5 minutes with 1M ethanolamine-HCl ( pH 8 . 5 ) . The coated biosensors were then equilibrated in PBS ( pH 6 ) for 10 minutes so as to wash away excess ethanolamine . Association between the antibodies and FcRn complex was evaluated by dipping the coated biosensors in 30 µg/mL FcRn complex in PBS ( pH 6 ) for 20 minutes . The biosensors were then dipped in PBS ( pH 6 ) for 20 minutes to evaluate dissociation . The FcRn-binding signals were normalized in the following manner: the signals obtained from FcRn-binding to IgG were divided by the signals obtained from total IgG on the biosensor . GPCMV-free , timed pregnant guinea pigs were inoculated by subcutaneous injection with 4×103 PFU of IVP8 stock ( 4 guinea pigs ) or media as a mock-infection control ( 3 guinea pigs ) at 21 days gestation . Guinea pigs were administered with 1968/GPFc antibody at 10 mg/kg at 1-day post-infection . Blood samples ( approximately 400 µl each ) were collected for ELISA analysis via the orbital route prior to dosing ( 0 h ) and at the following times after dosing with 1968/GPFc antibody: 0 . 5 hr , 8 hr , d1 , d3 , d7 , d14 , and d21 . The serum samples were diluted in PBS+0 . 5%BSA+0 . 25% CHAPS+5 mM EDTA+0 . 35M NaCl , +10 ppm Proclin+0 . 05% Tween 20 at pH 7 . 4 and analyzed by ELISA using soluble GPCMV gH/gL protein ( see ELISA section for details ) . Serum interference was evaluated with multiple conjugates from Jackson and Novus prior to sample evaluation with the Donkey anti-guinea pig IgG ( H&L ) HRP providing the best signal . The minimum dilution was found to be 1/400 with a limit of quantification at 0 . 62 ug/mL . Standard curve range is at 1–100 ng/mL . Samples were diluted starting at 1/400 and then serially diluted 1/3 for a total of 8 points . Dilutions that fell within the standard curve were averaged to give the final concentration . Sera from seronegative animals were used to determine background levels . Serum concentration profiles for each guinea pig were analyzed individually , and mean ( s . e . m . ) values for the PK parameters were reported . 15 GPCMV-free , timed pregnant guinea pigs were inoculated by subcutaneous injection with 4×103 PFU of IVP8 stock at 21 days gestation . 1968/GPFc and control anti-gp120 antibodies were administered I . P . to 7 and 8 guinea pigs , respectively . The first dose was given one day prior to infection and then twice per week for 3 weeks at 8 mg/kg dose for a total of 6 doses . At day 42 of gestation , guinea pigs were euthanized and GPCMV DNA copy number was determined in the maternal salivary glands and placenta by qPCR and fetal tissue and amniotic fluid by nested PCR ( see PCR sections below for more details ) . Maternal salivary glands ( parotid , submandibular and sublingual glands were combined ) , placenta , and fetal tissue ( the entire fetus except the head ) were homogenized with gentleMACS Dissociator ( MACS Miltenyi Biotec ) . Salivary gland homogenates were centrifuged to remove debris . 200 µl of the homogenized tissue or 200 µl of amniotic fluid was used for DNA isolation with Qiagen DNeasy blood and tissue kit . DNA samples were analyzed using real-time quantitative PCR or nested PCR . DNA from whole blood was purified using the QIAamp DNA blood kit . The quantitative PCR ( qPCR ) assay targeting the GPCMV GP83 gene for quantification of GPCMV DNA was performed using standard ABI TaqMan protocols . TaqMan primer probe sets were as described by Katano et al . [40] . The primers internal to GP83 were used for qPCR: GP83F , 5′-CGACGACGACGATGACGAAAAC , and GP83R , 5′-TCCTCGGTCTCAACGAAGGGTC with the addition of the FAM probe 5′-ATCCGAGTTAGGCAGCG . To normalize and/or obtain the GPCMV DNA copy numbers in a single cell , copy numbers of the guinea pig actin gene ( GenBank accession number AF508792 ) were determined by qPCR using primers 5′-TGGATCGGCGGCTCTATC and 5′-CATCGTACTCCTGCTTGCTGAT with the VIC probe 5′-CACTCTCCACCTTCC . As noted above in the Neutralization Assays section , when relative percentages of infection were determined for neutralization assays , multiplex qPCR was performed with both primer/probe sets . In contrast , single primer/probe sets were used when determining absolute copy number ( e . g . from guinea pig tissue samples ) . Both the GP83 and actin genes were cloned into TOPO vector ( Life Technologies ) and used for making a standard curves to determine absolute viral copy number . Genomic DNA was analyzed from fetal tissue and amniotic fluid using the following sets of primers specific for the GPCMV GP83 gene . For the first round of PCR the following primers were used: GP83-1 , 5′- CCAACGTTCTCGGCCTGACGTTA , and GP83-2 , TGGGTACGCCGTCGAACC ( targeting a 972 bp product ) . The qPCR primers described above were used for the second round PCR reaction yielding a 248-bp product . PCR products from the first round were diluted 1∶10 and 5 µl of this was used in the 2nd round of PCR . Reactions and thermocycling conditions followed standard protocol using Phusion high-fidelity DNA polymerase ( New England BioLabs ) except for annealing temperatures and cycle numbers , which were 65°C and 30 cycles for the first round , and 60°C and 20 cycles for the second round . PCR products were analyzed on 2% agarose gels . Negative and positive controls at both rounds were included . Two main statistical analyses were done: the first was using the “bridged” kinetics studies ( Figures 2 and 3 ) and the second was to analyze the difference in infection rates with and without a prophylactic at day 21 ( i . e . the protection study ) . Since both analyses were focused on the fetuses , we had to take into account that fetuses are nested within mothers , and as such , should not be counted independently . A common approach for this situation is a beta-binomial model , where within each dam the infections of the fetuses are binomial and the proportion of fetuses infected over all the dams comes from a beta distribution [41] . For the kinetics experiment we fit a model specifying infection rate as a function of takedown time and cohort to account for experimental variation across cohorts . In the protection experiment we used treatment and cohort as terms in the model . Other simple analyses done were to look at mortality in the dams , as that is an indicator of how virulent the virus is within a cohort and abortion rate between infected and unaffected animals . Virulence , as a function of dam mortality , is confounded with cohort , so could not be used to adjust our models further .
|
Human cytomegalovirus ( HCMV ) is the most common cause of congenital virus infection and causes developmental abnormalities , including hearing loss and developmental delay . Although there is no therapy for congenital HCMV disease , there is evidence from both human and animal studies that antibodies can have efficacy in this setting . Such studies have focused exclusively on polyclonal antibodies , in which the targets of protective antibodies are unknown . Guinea pigs have been used as a model of human maternal fetal transmission of infection because of similarities in placental anatomy between human and guinea pig . Furthermore , guinea pig CMV ( GPCMV ) has been demonstrated to cross the placenta and cause fetal infection and loss , similar to the effects of infection with HCMV . However , the kinetics of maternal and fetal infection in this model has not been carefully investigated . In this work , we have delineated the kinetics of maternal to fetal infection and found that congenital infection is rapid following maternal infection . Importantly , we demonstrate that a monoclonal antibody against a protein critical for viral entry protects pregnant guinea pigs against fetal infection . Thus , our studies may be informative for development of a therapeutic intervention to treat congenital HCMV infection in humans .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases",
"biology",
"and",
"life",
"sciences",
"microbiology",
"medicine",
"and",
"health",
"sciences"
] |
2014
|
A Neutralizing Anti-gH/gL Monoclonal Antibody Is Protective in the Guinea Pig Model of Congenital CMV Infection
|
Researchers have conjectured that eye movements during visual search are selected to minimize the number of saccades . The optimal Bayesian eye movement strategy minimizing saccades does not simply direct the eye to whichever location is judged most likely to contain the target but makes use of the entire retina as an information gathering device during each fixation . Here we show that human observers do not minimize the expected number of saccades in planning saccades in a simple visual search task composed of three tokens . In this task , the optimal eye movement strategy varied , depending on the spacing between tokens ( in the first experiment ) or the size of tokens ( in the second experiment ) , and changed abruptly once the separation or size surpassed a critical value . None of our observers changed strategy as a function of separation or size . Human performance fell far short of ideal , both qualitatively and quantitatively .
For many detection and discrimination tasks , performance decreases rapidly with increasing distance from the center of vision . Observers overcome this limitation by making discrete eye movements ( saccades ) as often as three times per second , in effect scanning the environment . Such serial scanning is not limited to humans or to the visual modality . It is commonly found whenever the sensory range is limited spatially but the sensors can be displaced . Examples include exploratory whisking by rats [1] echo-location by bats [2] , and haptic exploration by humans [3] . The pattern of eye movements depends on the observer's goals [4] , [5] , [6] , [7] . In visual search , for example , the observer is searching for a specified target within the visual field . Following each eye movement the visual system gains access to new information as a result of the most recent eye movement and must decide whether to terminate the search because the target has been located , to continue the search by planning a further eye movement , or to abandon the search . If the search is continued then a key question is , how does the visual system plan the next saccade given the visual information gathered so far ? Models of eye movement planning fall roughly into two categories . The first class , salience models , uses the current retinal image to assign a numerical measure called salience to each location in the retina [8] , [9] , [10] . Salience is often linked to physical measures such as luminance or local contrast . Salience models differ in how salience is computed and in how the visual system uses the salience map to plan the next saccade . Models of the second class , optimal statistical models , are designed to optimize a specified criterion [11] . These models take into account the visual sensitivity of the eye across the retina and make use of all of the information gained in past searches to plan a sequence of saccades that , for example , minimizes the expected number of saccades needed to locate the target [12] , [13] , [14] , [15] , [16] . The information gathered during initial fixation and with each successive saccade is a measure of the likelihood that the target is at each possible retinal location , a likelihood map ( Figure 1 ) . While the ability of such statistical models to predict eye movements behavior in natural scenes has been challenged [17] , [18] , [19] , [20] , [21] , [22] and alternative models have been proposed , in particular that of Tatler and colleagues [23] incorporating high level features , statistical models allow to model ideal ( optimal ) behavior and compare human performance to ideal [11] . The differences between salience models and statistical models are less than they first appear to be . Likelihood , for example , is arguably a candidate measure of salience . However , a major difference between the two classes of model is the rules for planning the next saccade . With salience models these rules are typically ad hoc , chosen to capture known features of human visual search . They usually propose that the next saccade go to the currently “most salient” location but with some mechanism inhibiting return to those that have already been searched ( inhibition of return [24] , [25] ) . The planning algorithms for statistical models , on the other hand , are dictated by the requirement that search be optimal by a previously specified criterion . The modeler typically has no further choices once visual sensitivity across the retina is measured and the criterion to be optimized is selected . Recently , Najemnik & Geisler [13] analyzed the performance that could be expected of a statistical model designed to minimize the number of saccades needed to locate a target . Given the current likelihood map , it is intuitively appealing to plan the next saccade to the location most likely to be the target , the maximum likelihood point denoted in Figure 1 . Najemnik & Geisler [13] demonstrate that the correct optimal strategy , minimizing the expected number of saccades to locate a target , typically aims at a location that need not coincide with . is the location that allows the visual system to best use its extra-foveal retinal sensitivity to evaluate multiple locations simultaneously as illustrated in Figure 1 and in the accompanying inset . The likelihood of a target at the location may be very low , as indicated in Figure 1 . Moreover , the optimal strategy also considers possible information gathered from future searches contingent on information gained from the current , much as a strong chess player thinks beyond the immediate consequences of his current move . Najemnik & Geisler [13] compared human performance in searching for a small Gabor patch in noise background to the predictions of an optimal statistical model and found qualitative agreement between the model and performance , at least in overall performance . One difficulty in comparing performance between human and model is that the stimuli are complex and it is difficult , to predict trial by trial , where the ideal observer should fixate . Here we present a simplified visual search task which allows us to test whether the visual system uses its extra-foveal sensitivity ( as Najemnik & Geisler [13] propose ) to minimize the number of saccades required to identify the target . In this task , the observer makes only one saccade per trial and we restrict the observer's possible choices of saccadic destinations to three . The observer must saccade to one of these three possible locations , marked by gray squares arranged horizontally , above or below his initial fixation ( Figure 2A ) . When the observer has completed the saccade , the target appears at either the left or right location , but never in the center . The trial is aborted if the observer tries to execute a second saccade . The target consists of a grey square and a small white dot that is either near the top of the square ( dot-up configuration ) or near the bottom ( dot-down configuration ) . The two configurations are shown in an inset in Figure 2A . The observer's task is to discriminate whether the target is dot-up or dot-down . He receives a small amount of money for each correct discrimination . The observer's probability of correct discrimination is determined by his retinal sensitivity function where ( eccentricity ) is the distance from the fovea to the target . For the discrimination task we employed , is a decreasing function of . We plot an example of versus for one observer in Figure 2B . The observer has only three possible choices of strategy . He may saccade to the leftmost token , the center token , or the rightmost . If the observer adopts the center strategy , then the separation between where the observer is fixated and the target on left or right is just the spacing between the locations ( denoted in the Figure 2A ) . The probability of correct discrimination is ( 1 ) If the observer adopts a side strategy then the separation will be either 0 ( if he has chosen the location where the target appears ) or ( if he has chosen the location on the side opposite to the location where the target appears ) . Since the target appears at the left or right location with equal probability , the observer's probability of correct discrimination is ( 2 ) In Figure 2C we plot and for the observer whose retinal sensitivity map is shown in Figure 2B . If the tokens are close together ( is small ) then the probability correct is close to 1 for both strategies . When the separation is between and , use of the center strategy would lead to higher probability correct . Beyond the point marked use of the side strategy would maximize expected probability correct . This critical value is determined by the observer's retinal sensitivity function . If the human observer is using his peripheral sensitivity to maximize the probability of correct discrimination , we would expect an abrupt change in strategy when the separation of center and side tokens exceeds . In Experiment 1 we first measured observers' retinal sensitivity functions . In the main part of the experiment , observers chose between center and side strategies as we varied separation over the range 8 to 24 degrees . Observers received a small monetary reward for each correct discrimination . We compared observers' choices of strategy ( center or side ) to the choice of strategy maximizing expected gain . The observer maximizing expected gain would pick the strategy , center or side , offering the larger probability correct in Equations 1 and 2 , switching strategy at the optimal switch point . In Experiment 2 we varied the size of the targets rather than distance to manipulate . Observers chose between the same array of tokens in Figure 2A but now the tokens varied in size . There is still an optimal point in where the observer should switch from a center strategy to a side strategy but now it is expressed in size . Each experiment consisted of three phases , sensitivity mapping , decision , and verification , illustrated in Figure 3 and described in the Methods section . In the sensitivity mapping phase , we measured sensitivity for the visual task for different eccentricities of targets ( Experiment 1 ) and for different sizes of targets at different eccentricities ( Experiment 2 ) . In the second ( decision ) phase , we tested human ability to select eye movements that maximize expected gain . In the last phase ( verification ) , we repeated the decision phase but forced the observer to make the saccade that our model ( see Methods section ) predicted would maximize expected gain . By doing so , we verified that , had they followed this strategy , they would have increased their expected gain to the maximum possible expected gain predicted by the model . In the separation and size experiments , we assumed that the target is always presented and we considered only the first saccade . If we were to modify the task slightly so that , although the grey squares appeared , the target configuration ( dot-up or dot-down ) was not presented on one half of the trials ( that is , all of the grey squares were uniform , without a marked configuration ) , then on trials where the observer fails to detect the location of the configuration after one saccade , he must make one or two additional saccades to determine if the target configuration is present at all and in what configuration . The strategy in our task which maximizes probability correct also minimizes the number of saccades needed to be correct in this modified task . In the experiments reported , we interleaved separations ( Experiment 1 ) and sizes ( Experiment 2 ) rather than presenting the same separation or size repeatedly in in a single experimental block . Observers could potentially learn the blocked task by simply trying different saccadic strategies and seeing which is more rewarding as a function of separation ( Experiment 1 ) or size ( Experiment 2 ) . In effect they learn to pair strategy and separation/size by reinforcement learning . But the prediction of the class of statistical models we consider is that observers will take into account their own retinal sensitivity in planning saccades without an extensive history of reinforcement [26] . Reinforcement learning plays no role in these theories .
The sensitivity mapping plots are presented in Figure 4A for Experiment 1 and in Figure 4B for Experiment 2 . The maps show the percentage of correct responses as a function of Eccentricity for Experiment 1 and Size for Experiment 2 . The results for each observer are used to predict the ideal observer performance in the decision phase .
We reported two experiments intended to determine whether observers correctly employed their extrafoveal retinal sensitivity to optimize visual search . Each experiment consisted of three phases , sensitivity mapping , decision , and verification . In the sensitivity mapping phase , we measured each observer's retinal sensitivity as a function of target eccentricity ( Experiment 1 ) and/or target size ( Experiment 2 ) . On each trial of the decision phase , observers first executed a saccade to one of three retinal locations , left , center or right ( Figure 2A ) . Following the saccade , a target would appear at either the left or right location but never in the center location . The observers then attempted to discriminate whether a small white dot within the target was near the top or bottom of the target . Their probability of success in discriminating depended on the location to which they had saccaded and the location at which the target appeared . Observers received a monetary reward for each correct discrimination and their challenge was to decide which location to saccade to so as to maximize their expected gain . We refer to their choices as a saccadic strategy . There were only three possible strategies , left , center , or right , and two of these strategies , left and right , were effectively equivalent ( see Methods ) . We refer to them collectively as the side strategy . A center strategy led to better discrimination for smaller eccentricities ( Experiment 1 ) or larger sizes ( Experiment 2 ) . Whether the target appeared on the left or right side , the small eccentricity ( large size ) meant that the observer could discriminate above chance while fixated at the center location . For large enough eccentricities in Experiment 1 , the center strategy resulted in performance near chance . In contrast , either side strategy resulted in better performance since , if the target appeared on the same side as the observer chose to saccade to , then he could readily discriminate it . This would occur on half the trials and on the remaining trials , when the target appeared on the side not chosen , the observer would be near chance in responding . Overall , the side strategy would lead to performance better than that expected with the center strategy . See Figure 2C . The same conclusion holds in Experiment 2 where we varied size . Consequently , as the experimenter increased the eccentricity of the side locations or decreased the size of the target , the observer optimizing expected gain or , equivalently , probability correct , should switch from a center strategy to either one of the side strategies at a specific optimal switch point . We used the data from the sensitivity mapping phase to predict the optimal switch point as a function of eccentricity or in size for each observer . We compared observers' choices of strategy to the choices predicted to maximize their probability correct in the discrimination task . None of our 6 observers switched strategy at the optimal point . All had evident , idiosyncratic biases , toward either the side or center strategies , but , most strikingly , they chose the center and side strategies equally often for all eccentricities and sizes of target . They did not adapt their strategy to the stimulus configuration at all . In a separate verification phase we reran the main part of the experiment but now indicating to the observer where to saccade on each trial , “forcing” the observer to adopt the saccadic strategy that our model predicted would maximize probability correct and expected gain . We found that observers' mean gain increased when they executed the strategy predicted to maximize expected gain and that their mean gain was in good agreement with the maximum expected gain predicted by the model . In summary , observers did not respond to variations in token separation and size at all . They apparently ignored the independent variable in each of the two experiments . We had expected that , for example , their performance might be qualitatively consistent with that of the ideal observer . That is , they might abruptly switch strategies at some separation ( size ) but not at the separation ( size ) that maximized expected gain . Or they might have been inconsistent in choosing strategies but only near the switch point so that probability of picking the side strategy might smoothly decrease from 1 to 0 instead of following the ideal step function . Neither of these outcomes occurred . We see no evidence that the visual system is sensitive to the factors we varied in the experiment . Humans sometimes do not make single saccades even when it is possible instead producing two or three saccades [31] . We considered the possibility that this particular aspect of our protocol is responsible for observed sub-optimality . An argument against this possibility is the ease with which our subjects adapted to the task . The ratio of excluded trials due to blinks or second saccade combined was between 8% and 24% of the trials ( 15% on average ) . If the observer's saccade did not arrive within one degree of a token , we terminated the trial . This occurred on between 2 . 4% and 34% of trials ( mean 17% ) , across observers . However , the horizontal and vertical standard deviations of saccades in our experiment ( see Figure S1 in the Text S1 ) are large compared to the one degree cutoff we imposed and , even if the observer attempted to saccade to the center of a token on every trial , many of the resulting saccades would fall outside the one degree limit . Hence , we cannot infer that a failure to saccade to within one degree of a token ( the criterion for success ) indicates that the observer intended to saccade anywhere other than to the token . In particular , there is no basis to conclude that normal eye movement planning and execution has been altered by the constraints we impose . Had we used a less stringent criterion for termination of a trial ( or imposed no criterion ) , observers might have chosen to saccade to a location away from any of the three tokens . Our analysis depends on knowing what strategies are available to the observer and which of these they chose . We also verified that the observer's distributions of saccades to each token were approximately centered on the token and not off to one side . The distributions of saccade endpoints for all observers are show in Figure S1 of the Text S1 as well as descriptive statistics for all the observers . If planning consumes cognitive resources then the choice of optimal plan should reflect the “cost” of these resources to the organism [32] . The key problem , though , is to develop experimental methods that allow us to demonstrate that these hidden costs are real and that they explain the observer's behavior . In conclusion , we find little evidence that observers correctly use their visual sensitivity outside the fovea to optimize visual search . Our results are in apparent conflict with the predictions of optimal statistical models discussed in the introduction [12] , [13] , [14] , [15] , [16] . Najemnik & Geisler [13] , for example , asked observers to locate a Gabor patch in a 1/f field of noise . They compared human performance to ideal performance minimizing the expected number of saccades to find the target . As we explained in the introduction , the strategy that maximizes expected gain and probability correct in our task also would minimize the number of saccades needed to correctly discriminate the target configuration . Our task is designed so that the visual system must have access to estimates of retinal sensitivity as a function of size or eccentricity in order to plan saccades that maximize expected gain . We , in effect , compared choice of saccade on each trial to the choice of saccade that would maximize expected gain , something we could do because of the simplicity of our design . The key predictions of Najemnik & Geisler's model are more difficult to match to human performance in their experiments . They , for example , predict the length of the first saccade and find that the distributions of lengths of first saccades are matched to that of the ideal . However , this does not imply that any particular saccade , triggered by a particular combination of signal and noise , is in itself optimal or even close . An alternative explanation for the results of Najemnik & Geisler [13] is that human visual search is based on simple heuristics analogous to those postulated in salience models . Tatler & Vincent [33] for example , presented compelling evidence that saccade selection could be better predicted by oculo-motor preferences than by visual information or task ( although they did not provide evidence of predictive power of these biases relative to chance ) Under this account , the visual system has heuristic preferences for saccades of certain lengths or possibly a tendency to saccade to the center of mass of clusters of objects in the periphery [34] , [35] . The second heuristic , under specific circumstances , might mimic selection of the optimal point in Figure 1 not because it is the saccade that minimizes the expected number of saccades but because it lies near the centroid of a cluster of items in the visual field . Such a heuristic-based approach may approximate ideal performance in some tasks while failing utterly in others . The experimenter who considers performance in a limited range of scenes may record behavior that approximates optimal but is in fact no more than a lucky coincidence of a heuristic rule and experimental conditions . Such “apparent optimality” is not rare in behavioral studies of animals [36] or humans [37] . And , since the stimuli of Najemnik & Geisler [13] were chosen to mimic the statistical properties of natural scenes , it is not surprising that application of visual heuristics lead to good performance in such scenes . If human saccade decisions are based on such heuristics rather than on a computation that requires knowledge of visual sensitivity maps , we would expect a failure of adjustment when one's visual sensitivity map is changed . In fact , when observers' foveae were artificially shifted with gaze-contingent techniques , their performances in visual search were significantly worse than predicted by the ideal-observer model [38] . In contrast , we designed our simple task so that the visual system can only succeed if it has access to estimates of visual sensitivity for the range of sizes and eccentricities we considered . We compared human performance to optimal on a trial by trial basis . We conjecture that observers failed in our task because it is not well matched to the collection of visual heuristics that guide saccadic selection .
Experiments were programmed in C++ using Microsoft DirectX APIs on a Pentium 3 computer running Windows XP . Stimuli were displayed on a 19-inch Sony Trinitron Multiscan G500 monitor run at a frame rate of 100 Hz with 1280×1024 resolution in pixels . A forehead bar and chinrest were used to help the observer maintain a viewing distance of 57 cm . At that distance , the full display subtended 40 . 4°×30 . 3° . The observer viewed the display binocularly . Eye movements were recorded using an Eye Link II ( SR Research , Toronto , Canada ) sampling eye position at 500 Hz . The subjects were NYU undergraduate students . Four subjects participated in the Experiment 1 ( 3 female ) and two in Experiment 2 ( 1 female ) . They were unaware of the purpose of the experiment and all had normal or corrected-to-normal vision . Stimuli were presented against a uniform gray background ( 50% white ) . The target configuration , represented in the inset in Figure 2 , consisted of a light gray square with a superimposed light gray dot at either the top ( dot-up configuration ) or at the bottom ( dot-down configuration ) . The tokens subtended 1° of visual angle in Experiment 1 and between 0 . 6° to 1 . 8° of visual angle in Experiment 2 . The observer's task was to report whether the target was dot-up or dot-down . Observers responded by rotating the mouse wheel in one direction corresponding to dot-up , the other corresponding to dot-down . Observers were rewarded for correct responses and they were aware that they would be rewarded . Observers were instructed to reply as accurately as possible and no time was imposed on their response . They were not given any feedback regarding their response . The maximum reward was $20 . Each experiment comprised three phases , sensitivity mapping , decision and verification . We ran two experiments , in Experiment 1 we varied only the separation and in Experiment 2 , only the size . The different separations in Experiment 1 were randomly interleaved , the different sizes in Experiment 2 were also randomly interleaved . As explained in the Introduction , there is a given separation ( Experiment 1 ) or size ( Experiment 2 ) at which observers should switch strategy . We call the optimal switch point and , in Experiment 1 , it is defined as the separation between the side tokens for which ( 5 ) The right hand side of the equation describes the observer's performance when he has chosen to saccade to one of the side tokens . On half the trials , the target will appear at that side location and he will discriminate correctly with probability ( he is fixating the target ) . On the other half the trials , the target will appear on the other side , a distance from fixation . He will discriminate correctly with probability . The overall probability of correct discrimination is the right hand side of Equation 3 . The left hand side is the performance expected with a center strategy . Whether the target appears on left or right , it is a distance from fixation and the observer discriminates correctly with probability . The switch point is the point at which the two strategies lead to equal discrimination performance . For eccentricities with , saccading to the center square results in a higher probability of correct classification . For with , saccading to either of the side tokens leads to better performance . We derived a similar equation for Experiment 2 but now in terms of target size . The optimal switch point is defined by ( 6 ) with denoting the sensitivity mapping function for each observer and denotes the sensitivity function as a function of size for eccentricity . At this point , both strategies have the same probability of success . The ideal observer that maximizes expected probability correct will switch strategy precisely at and . We estimated and for each observer in each experiment using Equations 5 and 6 and numerical optimization . We also verified that the optimal point is unique . The and for each observer are shown together with the results in the next section . The sensitivity functions and could also be estimated using the data from the decision or verification phase . We used the sensitivity function derived from the sensitivity mapping phase in the analysis reported in the main text . Using the data from either of the other two phases only led to small changes in estimated optimal switch point that do not affect our conclusions . We report those in Figure S3 in the Text S1 .
|
Vision is most sensitive to fine detail at the center of gaze ( the fovea ) . We typically move our eyes several times a second to build up an accurate picture of the world around us and find objects of interest . Very recently , researchers have developed models of how a visual system like ours could search a scene for a specific target with the smallest possible number of eye fixations . In two experiments , we tested the assumptions underlying such models . We set up visual “games” in which observers were rewarded for their performance in moving their eyes once to recognize simple targets . To do well ( earn the maximum possible reward ) , observers had to move their eyes according to the predictions of recent models of eye movement . We found that our observers failed to choose optimal eye movement strategies and failed to maximize their potential winnings . Our results suggest a simpler picture of eye movement selection , driven by a few simple heuristic rules that lead to good but not optimal performance in everyday tasks .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"cognitive",
"neuroscience",
"psychology",
"social",
"and",
"behavioral",
"sciences",
"biology",
"neuroscience"
] |
2012
|
Human Visual Search Does Not Maximize the Post-Saccadic Probability of Identifying Targets
|
The winter seasonality of influenza A virus in temperate climates is one of the most widely recognized , yet least understood , epidemiological patterns in infectious disease . Central to understanding what drives the seasonal emergence of this important human pathogen is determining what becomes of the virus during the non-epidemic summer months . Herein , we take a step towards elucidating the seasonal emergence of influenza virus by determining the evolutionary relationship between populations of influenza A virus sampled from opposite hemispheres . We conducted a phylogenetic analysis of 487 complete genomes of human influenza A/H3N2 viruses collected between 1999 and 2005 from Australia and New Zealand in the southern hemisphere , and a representative sub-sample of viral genome sequences from 413 isolates collected in New York state , United States , representing the northern hemisphere . We show that even in areas as relatively geographically isolated as New Zealand's South Island and Western Australia , global viral migration contributes significantly to the seasonal emergence of influenza A epidemics , and that this migration has no clear directional pattern . These observations run counter to suggestions that local epidemics are triggered by the climate-driven reactivation of influenza viruses that remain latent within hosts between seasons or transmit at low efficiency between seasons . However , a complete understanding of the seasonal movements of influenza A virus will require greatly expanded global surveillance , particularly of tropical regions where the virus circulates year-round , and during non-epidemic periods in temperate climate areas .
Influenza A virus is able to persistently re-infect human populations by continually evading host immunity through the rapid evolution of surface antigens ( “antigenic drift” ) [1] . Influenza virus epidemics strike temperate latitudes of the world each winter , from November to March in the northern hemisphere and from May to September in the southern hemisphere [2] . In the United States alone , these influenza epidemics are associated with an annual average of 36 , 000 human deaths [3] and 226 , 000 hospitalizations [4]; globally , the virus is associated with as many as half a million annual deaths [5] . While rapid antigenic change is a hallmark of influenza evolution , recent studies have failed to detect antigenic drift over an epidemic season , suggesting that important evolutionary processes may occur during non-epidemic periods , either locally or perhaps elsewhere [6–8] . However , surveillance during non-epidemic periods is not conducted routinely by the network of World Health Organization influenza reference centers [9] and , consequently , little is known about how and where the virus persists in the human population in between winter epidemics at low levels . A key question is therefore whether the virus remains locally within its host population in between epidemics , perhaps persisting within hosts in a latent state [10] , or whether the virus migrates afar to other reservoirs , such as the tropics , and is later reintroduced . Although influenza virus has long been regarded a “cold-weather” pathogen due to its marked winter epidemics in temperate zones , recent studies show that tropical regions experience significant year-round influenza virus activity [11] . In theory , such a “tropical belt” could serve as a year-round reservoir that harbors endemic populations of influenza virus that seasonally reintroduce viral isolates into temperate zones to trigger new epidemics [12 , 13] . Whereas population crashes at the end of seasonal epidemics create severe evolutionary bottlenecks that limit genetic diversity , tropical zones may function as permanent mixing pools for viruses from around the world . Historically , Southeast Asia has been considered a potential epicenter for emergence of pandemic viruses due to the proximity with which humans live with their domestic animals [14] . However , abundant data from these regions is currently unavailable , so the origins of influenza pandemics and epidemics remain unclear . Given the ease and speed with which the influenza virus is thought to spread between humans , it is generally accepted that global chains of direct person-to-person transmission are sufficient to maintain the influenza virus in the human population [15] . However , a complete understanding of how the influenza virus transmits between humans is lacking [16] , and whether human-to-human spread alone accounts for the seasonal emergence of epidemics has been questioned [17] . The simultaneous appearance of influenza outbreaks separated by large longitudinal distances , as well as sporadic influenza cases during summer months , suggests that the virus may instead already be “seeded” and somehow reactivated by environmental stimuli . Thus , the alternating pattern of northern and southern hemisphere bi-epidemics could , in principle , also result from opposite climatic forces independently reactivating viral activity in these two hemispheres at alternating six-month intervals . Hence , instead of continually migrating across the equator , separate viral populations could persist locally in an asymptomatic latent state over the summer months until climatic stimuli sufficiently increase host susceptibility and/or viral transmissibility to induce another epidemic . However , hypotheses of how climatic change may directly or indirectly influence viral activity and/or host susceptibility remain largely untested [18 , 19] . Crucially , some theories for influenza seasonality produce testable phylogenetic hypotheses . On one hand , if influenza A virus persists locally over the summer in a latent state , then isolates sampled over multiple seasons from a single locality would cluster together on a phylogenetic tree , separate from isolates from other geographic regions ( Figure 1 ) . Alternatively , if the virus did not evolve in situ between epidemic seasons , but rather traveled globally between epidemics , then the resulting phylogeny would show extensive intermixing of isolates from different localities . To determine whether influenza virus migrates away between the northern and southern hemispheres during non-epidemic summer months , or remains there latently , we conducted an extensive phylogenetic analysis of 399 whole-genome A/H3N2 influenza A viruses sampled from New Zealand ( most commonly Canterbury , South Island ) from 2000 to 2005 ( six seasons ) , 88 viral genome sequences from Australia ( most commonly Western Australia ) from 1999 to 2005 ( seven seasons ) , along with a carefully selected sub-sample of 52 isolates that are representative of the clades present in a larger sample of 413 viruses from New York state , United States , collected between 1998 and 2005 and analyzed previously [7] . Given the relative geographic isolation and low population densities of New Zealand's South Island and Western Australia , as well as sampling limitations , this analysis provides a conservative estimate of the extent of cross-hemisphere migration occurring during this time period .
Populations of A/H3N2 influenza virus in Australia and New Zealand from 1999 to 2005 exhibit extensive genetic diversity across the entire genome ( Figures 2–4 ) , comparable to the diversity observed previously in New York state [7] ( in all phylogenies , clades from Australia are shaded blue , those from New Zealand in green , those from New York state in orange , and global isolates in pink ) . In particular , multiple viral clades co-circulate during each influenza season in New Zealand and Australia , defined as clusters of isolates with high bootstrap support ( >70% ) , or which are separated by exceptionally long branches ( Table 1 ) . As with the New York state data , these viruses were collected in the context of seasonal surveillance efforts in Australia and New Zealand and therefore likely provide a representative sample of the overall genetic composition of the viral populations . On the neuraminidase ( NA ) tree ( Figure 2 ) , viruses from New Zealand and Australia fall into at least 15 distinct clades , some of which appear in multiple seasons: three clades circulated in the 1999 season ( clades v , vi , and ix; when data was only available from Australia ) , five in 2000 ( clades i , ii , iii , iv , viii ) , four in 2001 ( clades i , iv , v , g ) , one in 2002 ( clade a ) , one in 2003 ( clade b; when a new reassortant virus predominated ) , two in 2004 ( clades A and c ) , and three in 2005 ( clades A , B , E ) . Clade lettering reflects the three main sections of the phylogeny , based on topology and time: clades A–E contain viral isolates from 2003 to 2005 that fall within section I of the tree ( large pink rectangle in upper portion of tree ) ; clades a–e contain isolates from 2001–2003 that fall within section II ( large light green rectangle in middle portion of tree ) ; clades i–xiii contain isolates from 1998 to 2000 that fall outside of sections I and II . The hemagglutinin ( HA ) tree ( Figure 3 ) contains at least 15 clades from New Zealand and Australia: two circulated in 1999 ( clades v and x ) , four in 2000 ( clades i , ii , iii , iv ) , four in 2001 ( clades g , i , iv , xiii ) , two in 2002 ( clades h and j ) , one in 2003 ( clade b ) , two in 2004 ( clades A and c ) , and three in 2005 ( clades A , B , E ) ( Figure 3 ) . Finally , the phylogeny of the concatenated six non-surface glycoprotein segments ( PB2 , PB1 , PA , NP , M1 , NS1 ) contains the largest number of clades , presumably because this larger data set ( 9 , 636 bp ) provides the greatest resolution . Southern hemisphere isolates form at least 18 clades on this phylogeny: four were present in 1999 ( clades v , vi , x , xii ) , three in 2000 ( clades i/ii , iii , iv ) , four in 2001 ( clades g , i/ii , iv , xiii ) , three in 2002 ( clades a , k , l ) , one in 2003 ( clade b ) , three in 2004 ( clades A , F , c ) , and three in 2005 ( clades A , B , E ) ( Figure 4 ) . Differences in the number of clades among segments may also be indicative of reassortment , especially involving the HA gene , as previously demonstrated in New York state [7] . Indeed , several major reassortment events are immediately evident from topological incongruities among these three phylogenies . On the HA tree , clades b , c , and e fall in section I along with isolates from 2004 and 2005 , while these clades fall in section II amidst 2002 and 2003 isolates in the NA and concatenated six non-surface glycoprotein phylogenies ( Figures 2–4 ) . Clade c also falls in section I on the PB2 phylogeny , showing a similar reassortment pattern as HA . However , aside from this lone reassortment event , the phylogenies of the six non-surface glycoprotein segments are very similar , enabling us to study them as a single concatenated entity ( phylogenetic trees for individual segments are provided as Figures S1–S6 ) . Isolates from the 52 representative New York state genomes are clearly interspersed with southern hemisphere isolates throughout our phylogenetic trees ( Figures 2–4; Table 1 ) , indicating that these populations regularly intermix as a result of cross-hemisphere migration . However , patterns of viral intermixing are both variable and complex , as clades from all three phylogenies contain an array of different combinations of viral populations from New Zealand , Australia , and/or New York state . For example , the phylogeny of the concatenated six non-surface glycoproteins contains a relatively even mix of mono-hemisphere and bi-hemisphere clades ( Table 1 ) , with five New York state–only clades , seven southern hemisphere–only clades , six clades that contain New York state isolates and isolates from only one of the southern hemisphere countries , and three clades that contain isolates from all three countries . Thus , on an annual basis some , but not all , viral populations mix with other populations from the same and/or opposite hemisphere , and this number is likely to increase with additional sampling . Indeed , a majority ( nine ) of the 16 clades containing southern hemisphere isolates also contained northern hemisphere isolates , suggesting widespread viral traffic across the equator ( Table 1 ) . Migration between Australia and New Zealand is also extensive , as almost all clades containing isolates from New Zealand also contained Australian isolates , and vice versa , except for the 1999 season , for which no New Zealand isolates were available . Further , these figures are likely to be underestimates , as mono-national and mono-hemisphere clades may be at frequencies too low to be detected in the genome collections currently available . Alternatively , clades could also have originated in areas not sampled in our study , such as tropical regions , where influenza viruses typically circulate year-round [12] . Strikingly , even for those viral clades that do not exhibit cross-hemisphere migration , there is very little evidence for in situ evolution within specific localities . For example , in no case on any phylogeny are clades of A/H3N2 from New York state directly linked over multiple seasons . Rather , New Zealand and Australia viruses are always interspersed among New York state clades from different seasons , indicating that they are not evolving in geographic isolation across seasons . Most clades from New Zealand and Australia show equivalent patterns of discontinuous evolution , with very few clusters of southern hemisphere clades that are not separated by New York state viruses ( although in situ evolution cannot be ruled out between a few 2004 and 2005 clades without additional sampling ) . Thus , even in relatively isolated areas of New Zealand and Australia , viruses do not regularly evolve in geographic isolation . Rather , evolution appears to be shaped by frequent cross-hemisphere migration and recurrent reintroduction . Importantly , our phylogenetic analysis suggests that seasonal migration occurs from the northern hemisphere to the southern hemisphere , as well as south-to-north . Although inferring the direction of viral migration is in part dependent on sample composition , definitive evidence of a migration event are clades containing a single population of northern hemisphere viruses and a single population of southern hemisphere viruses supported by a high level of bootstrap support ( >70% ) . Because winter influenza seasons alternate by six-month intervals between the northern and southern hemispheres , it was also possible to determine , within the confines of sampling , the timescale and hence direction of cross-hemisphere migration . The inferred directionality of 11 cases of definitive cross-hemisphere migration evident on the HA , NA , and non-surface glycoprotein phylogenies are summarized in Table 2 . Of these , all but two involve the concatenated non-surface glycoproteins , which provide a more reliable phylogeny as previously described . Eight of these migration events occur in a north-to-south direction , versus three in a south-to-north direction , suggesting that viruses may migrate more frequently from New York state to the southern hemisphere than in the opposite direction , although this will need to be confirmed with larger sample sizes . For example , on the concatenated gene tree , viral isolates from the 2003–2004 season in New York state form a single well-supported phylogenetic clade ( clade b , 100% bootstrap support ) with viruses from 2003 from New Zealand and Australia ( Figure 4 ) . Since the 2003 New Zealand and Australia viruses predate the 2003–2004 New York state viruses ( i . e . , the northern hemisphere winter ) , we infer that the lineage that gave rise to these southern hemisphere viruses migrated northward to infect New York state between the 2003 southern hemisphere winter ( May to October ) and the 2003–2004 winter in New York state . It is also notable that cross-hemisphere migration does not follow any clear pattern . In addition to occurring in both north-to-south and south-to-north directions , migration events also appear to involve minor clades as frequently as major clades , assuming that our study sample is generally representative of the viral population structure ( Table 2 ) . Furthermore , the populations of southern hemisphere viruses that migrate northward are a mix of compositions , including some isolates only from New Zealand , some only from Australia , and populations with a mix of isolates from both countries ( Tables 1 and 2 ) . Finally , two clades contain global strains , but because the dates of these global isolates are not recorded , it is impossible to accurately determine the direction of migration events . The 11 cases presented in Table 2 represent only the strongest examples of cross-hemisphere migration , using the strictest criteria to infer migration events from the phylogenetic data . Relaxing this stringency allows for the possibility of greater bi-directional cross-hemisphere migration , especially involving clades containing more than one viral population from the southern hemisphere . Although the direction of migration is less certain when populations of viruses from three geographical regions are present , the relative frequency of migration observed under the more stringent criteria suggests that cross-hemisphere migration likely operates in these cases as well . Finally , while it is likely that a more intensive sampling regime will increase clade diversity , and in doing so affect the inference of the direction of migration , the complexity of the patterns observed strongly argues for frequent bi-directional migration .
Our large-scale phylogenetic analysis of A/H3N2 influenza virus populations from opposite geographic hemispheres provides evidence for regular bi-directional cross-hemisphere viral migration between seasons , even among localities as distantly separated as New York state and Australia and as relatively geographically isolated as New Zealand's South Island . Multiple genetic variants of influenza virus co-circulate each season , even in geographically remote areas , and many of these viral clades are more closely related to isolates from the opposite hemisphere than to isolates from either the previous or following season in the same location . Thus , viral populations do not appear to “over-summer” locally , where they would evolve in situ and give rise to the next season's epidemic . Rather , cycles of viral migration and recurrent introduction have clearly played a significant role in generating the genetic diversity that characterizes influenza A virus in both hemispheres . Importantly , given the sample composition of our sequence data set , the extent of cross-hemisphere migration observed here undoubtedly represents a conservative estimate . Hence , including data from more populated areas could only reveal more instances of cross-hemisphere migration . In addition , our finding that the virus migrates globally between epidemics and is reintroduced is clearly compatible with tropical regions , including Southeast Asia , playing a key role in the genesis of new clades and the global spread of these novel influenza virus variants . Thus , while limitations in global genome sampling necessarily means that the current study is directed toward testing hypotheses of viral migration versus latency , equivalent data from tropical regions would undoubtedly enable us to conduct a more refined analysis of global migration patterns and their determinants . Specifically , if tropical regions serve as year-long influenza reservoirs , we would expect to observe phylogenies in which tropical isolates display the greatest genetic diversity and are positioned basal to viruses sampled from temperate regions . Consequently , complete genome sampling from tropical regions where influenza viruses circulate year-round , including a record of the precise date of collection , is of key importance for understanding the global epidemiology of the influenza virus . Notably , the viral migration we observe does not appear to follow any clear pattern , but rather occurs in all directions , involves all genes , and involves clades of all sizes and geographic compositions . This argues against a role of immune selection in determining which viral clades are able to migrate among localities , although it does not preclude a role for natural selection as the sieve that determines which clades are able to survive in specific host populations . Similarly , the observation that migration patterns vary to some extent among the HA , NA , and concatenated non-surface glycoproteins must reflect the effect of widespread genomic reassortment [7 , 20] . Frequent reassortment complicates the analysis of migration patterns , as individual viruses can carry genomic segments with differing phylogenetic , and hence geographical , histories . Consequently , the analysis of migration patterns based on single gene segments may paint a misleading picture . Although the transmission of the influenza virus through population movements has been studied extensively , particularly for the spread of pandemic isolates across the globe by air travel [21 , 22] , neither the routes nor the mechanisms of the virus's geographical spread have been fully resolved . Several recent studies have used empirical data to investigate the role of population movements on the spatial diffusion of seasonal epidemics , including an intricate analysis of the regional spread of influenza epidemics across the United States , which was strongly correlated with adult workflow movements [23] . A previous epidemiological study comparing the synchronicity with respect to timing of influenza epidemics between the United States , France , and Australia suggested that the inter-hemispheric circulation of epidemics follows an irregular pathway , with recurrent changes in the leading hemisphere [24] , in accordance with the phylogenetic analysis presented here . More fine-scaled analyses of discrete viral populations have shown that frequent introduction of “foreign” viruses significantly impacts the viral population structure and geographic spread at local levels . For example , the rapid timescale of global mixing of influenza drowns out any impact of local heterogeneities on the spread of the epidemics through France [25] . Similarly , the seasonal importation of multiple global isolates appears to be a greater contributor to the genetic diversity of the influenza virus population in New York state from 1997 to 2005 than local in situ evolution [7] . While our findings confirm that human population movements play a role in introducing new viral variants at the start of an epidemic , some aspect of climate is clearly of importance in triggering epidemics . Additional research is required to define how human susceptibility to infection and viral transmissibility fluctuate under varying climate conditions and why influenza virus is absent in summer in temperate climates but exists year-round in tropical zones . Although the underlying cause of the seasonality of the influenza virus remains uncertain , even in reservoir avian species [26] , our findings illustrate the critical importance of expanding surveillance to elucidate the geographical movements and evolution of this virus throughout its entire annual cycle . The traditional focus on epidemic influenza may detract from the equally important epidemiological question of why influenza A virus does not circulate in humans for so many months of the year in temperate areas , especially given its apparent ability to infect humans in tropical areas year-round . Attempts to predict , model , or contain the spread of the influenza virus require a unified understanding of how the virus's spatial-temporal dynamics , antigenic evolution , and seasonal emergence interrelate [27] . Although this study is limited to only the three countries for which we have extensive data , our analysis exemplifies the capacity of phylogenetic analysis to elucidate challenging epidemiological questions by providing a level of finer resolution .
All influenza A ( H3N2 ) virus complete genome sequence data were collected from the National Institute of Allergy and Infectious Disease's Influenza Genome Sequencing Project ( http://www . niaid . nih . gov/dmid/genomes/mscs/influenza . htm ) for the period 1998–2005 [28] . Influenza A/H3N2 viruses were sampled by a network of participating general practitioners . Viruses from all 11 regions in New York state were collected by the Virus Reference and Surveillance Laboratory at the Wadsworth Center , New York State Department of Health . Influenza viruses from both the North and South Islands of New Zealand were collected by Canterbury Health Laboratories in Christchurch , New Zealand . In Australia , viruses from Western Australia were collected by PathWest Laboratory Medicine , Western Australia; viruses from New South Wales were collected by the Prince of Wales Hospital , New South Wales; viruses from South Australia were collected by the Institute of Medical and Veterinary Sciences , South Australia; and viruses from Queensland were collected by the Queensland Health Science Services , Queensland . All sequence data were downloaded from the National Center for Biotechnology Information ( NCBI ) Influenza Virus Resource ( http://www . ncbi . nlm . nih . gov/genomes/FLU/FLU . html ) . For Australia , 88 genome sequences from the 1999–2005 seasons were compiled , while for New Zealand , 399 genome sequences A/H3N2 sequences from the 2000–2005 seasons were collected . For New York state , United States , 52 phylogenetically representative genome sequences from the 1998–1999 to 2004–2005 seasons were carefully selected from a larger data set of 413 sequences from 1997–2005 analyzed previously [7] ( excluding 2000–2001 , for which few H3N2 sequences were available in an H1N1-dominant season ) . GenBank accession numbers for all sequences used in this study are listed in Table S1 . Sequence alignments were manually constructed for the major coding regions of each of the eight genomic segments . In addition to alignments for the HA ( 1 , 698 bp ) and NA ( 1 , 407 bp ) , an alignment of the concatenated six non-surface glycoproteins segments ( PB2 , PB1 , PA , NP , M1 , NS1 ) was also compiled ( 9 , 636 bp ) , as these are expected to evolve differently from the HA and NA surface glycoproteins . Because the minor M2 and NS2 proteins are involved in overlapping reading frames , they were excluded from the analysis . Initial phylogenetic trees were inferred for sequences of the HA , NA , and concatenated non-surface glycoproteins from New York state , New Zealand , and Australia under the HKY85 ( Hasegawa-Kishino-Yano ) model of nucleotide substitution using the Neighbor-Joining ( NJ ) method available in PAUP* [29] . Due to the very large size of all data sets , and the provisional nature of the analysis , the nearest-neighbor-interchange branch-swapping method was employed in this case . To assess the robustness of individual nodes on these phylogenetic trees , we performed a bootstrap resampling analysis ( 1 , 000 replications ) using the NJ method . From these three starting phylogenetic trees , “major” clades ( which contained the majority of isolates from a season ) and “minor” clades of genetically related viruses were identified by exceptionally long branch lengths and/or high bootstrap values ( >70% ) . A subset of sequences for the concatenated non-surface glycoproteins was constructed with 51 sequences from New Zealand , 45 sequences from Australia , and 52 from New York state ( see above ) for a total data set of 148 sequences . For the HA gene , these 148 isolates were placed in a more global context with the addition of 13 genetically unique HA sequences sampled from this time period available on GenBank , to produce a total of 161 HA sequences . Likewise , 22 global NA sequences were combined with the original 148 from New York state , New Zealand , and Australia for a total of 170 NA sequences . Maximum likelihood ( ML ) phylogenetic trees were then inferred using the PAUP* package [29] for these three new data sets: 161 HA sequences , 170 NA sequences , 148 concatenated sequences . ML trees were also inferred for each of the six non-surface glycoprotein segments to ensure that all exhibit similar tree topologies ( Figures S1–S6 ) . In each case , the best-fit model of nucleotide substitution was identified by MODELTEST [30] as the general reversible GTR+I+Γ4 model , with the frequency of each substitution type , proportion of invariant sites ( I ) , and the gamma distribution ( Γ ) of among-site rate variation with four rate categories ( Γ4 ) estimated from the empirical data . In all cases tree bisection-reconnection branch-swapping was utilized to determine the optimal tree . Finally , a bootstrap resampling process ( 1 , 000 replications ) using the NJ method was used to assess the robustness of individual nodes on the phylogeny , incorporating the ML substitution model . The analysis of the frequency and directionality of migration was undertaken through a visual inspection of the topological position of individual clades on each tree and in consideration of their time of sampling . Although more quantitative methods for determining migration patterns from gene sequence data have been established , particularly those based on parsimony reconstructions of changes in character state ( i . e . , geographical locality ) [31] , these were considered inappropriate for the current study because they ignore the temporal structure of the influenza virus genome sequence data . Specifically , we reasonably assume that older sampled clades give rise to younger sampled clades if they fall basal to them on phylogenetic trees .
|
The winter seasonality of influenza A virus in temperate climates is one of the most puzzling epidemiological patterns in infectious disease . To help resolve the issue of influenza seasonality , we studied , using viral genome sequence data , the patterns of global migration of influenza A virus , particularly between the northern and southern hemispheres . A phylogenetic analysis of approximately 900 complete genomes of the H3N2 subtype of human influenza A virus sampled from New Zealand and Australia ( southern hemisphere ) , and New York state , United States ( northern hemisphere ) , revealed that cross-hemisphere migration frequently occurs in both directions and involves multiple viral strains . Such global viral traffic therefore contributes significantly to the introduction of new influenza epidemics in both northern and southern hemispheres . These results also show that influenza A virus migrates afar during non-epidemic periods , rather than persisting locally at low levels during the influenza “off-season” . However , although this represents the largest and first bihemisphere study of its kind to our knowledge , the results highlight the need for sampling from tropical regions and during non-epidemic periods in temperate areas . Studies of this kind are critical to fully understand the geographical dispersal of influenza A virus and the role of climate in triggering seasonal epidemics .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"viruses",
"infectious",
"diseases",
"ecology",
"virology",
"evolutionary",
"biology"
] |
2007
|
Phylogenetic Analysis Reveals the Global Migration of Seasonal Influenza A Viruses
|
Yeast biofilms are complex multicellular structures , in which the cells are well protected against drugs and other treatments and thus highly resistant to antifungal therapies . Colony biofilms represent an ideal system for studying molecular mechanisms and regulations involved in development and internal organization of biofilm structure as well as those that are involved in fungal domestication . We have identified here antagonistic functional interactions between transcriptional regulators Cyc8p and Tup1p that modulate the life-style of natural S . cerevisiae strains between biofilm and domesticated mode . Herein , strains with different levels of Cyc8p and Tup1p regulators were constructed , analyzed for processes involved in colony biofilm development and used in the identification of modes of regulation of Flo11p , a key adhesin in biofilm formation . Our data show that Tup1p and Cyc8p regulate biofilm formation in the opposite manner , being positive and negative regulators of colony complexity , cell-cell interaction and adhesion to surfaces . Notably , in-depth analysis of regulation of expression of Flo11p adhesin revealed that Cyc8p itself is the key repressor of FLO11 expression , whereas Tup1p counteracts Cyc8p’s repressive function and , in addition , counters Flo11p degradation by an extracellular protease . Interestingly , the opposing actions of Tup1p and Cyc8p concern processes crucial to the biofilm mode of yeast multicellularity , whereas other multicellular processes such as cell flocculation are co-repressed by both regulators . This study provides insight into the mechanisms regulating complexity of the biofilm lifestyle of yeast grown on semisolid surfaces .
In nature , microorganisms preferentially live within multicellular communities such as different types of biofilms and colonies [1–5] . Yeast biofilms are complex uniquely organized structures , in which cells are protected from hostile environments , including antifungals , host immune systems , and starvation . Active multidrug resistance transporters and protective extracellular matrix that are produced by subpopulations of differentiated cells in colony biofilms formed by wild Saccharomyces cerevisiae strains [6] , contribute to this protection . Despite the identification of various genes and processes ( including chromosome reorganization ) involved in formation of colony biofilm structure [1 , 7 , 8] , many of these processes seem to be specific to particular wild strains . An exception is the Flo11p cell wall adhesin , a key protein involved in several developmental processes including cell adhesion [9] and the formation of colony biofilms [10] , flor biofilms [11] , and mats [12] . Deletion of FLO11 results in the formation of smooth colonies in various non-isogenic wild strains isolated from different habitats [13] as well as in Σ1278-derived strains [10 , 14] . FLO11 mRNA levels are elevated in colony biofilms and lowered after phenotypic switching called domestication , during which cells are reprogrammed to form smooth colonies similar to colonies of laboratory strains , in which key features of biofilm-life style are switched off [13 , 15] . The FLO11 promoter extends to 3 kb and contains at least four upstream activation sequences and nine elements involved in repression [16] . Thus , FLO11 gene expression integrates signals from diverse signaling cascades , including the Ras-cyclic AMP-dependent kinase , mitogen-activated protein kinase ( which controls filamentous growth ) and the main glucose repression pathways . These pathways positively or negatively regulate FLO11 expression in accordance with growth stage and nutritional conditions [1 , 16–18] . The expression of FLO11 is also controlled by epigenetic mechanisms , including histone deacetylation , chromatin remodeling , non-coding RNAs and prion formation [1 , 19–23] . The Cyc8p ( Ssn6p ) -Tup1p complex is mostly known to function as a transcriptional co-repressor that is conserved in eukaryotic organisms including mammals [24] . Four molecules of Tup1p in concert with one molecule of Cyc8p form this complex [25] with a tendency to oligomerize [26] , which regulates hundreds of S . cerevisiae genes involved in diverse pathways , such as glucose , starch and oxygen utilization , the response to osmotic stress , DNA repair , mating , sporulation , meiosis and flocculation [24 , 27 , 28] . Cyc8p-Tup1p does not bind directly to DNA but is brought to promoters via interactions with sequence-specific regulatory binding proteins , which coordinate the expression of specific subsets of genes [24] . Some data indicate that Cyc8p may play a more direct role in the repression [29] . Cyc8p-Tup1p can interact with Mig1p and Nrg1p , which bind to the promoters of glucose-repressed genes , such as FLO11 , in the presence of glucose [17 , 30] . Cyc8p-Tup1p can also act as a transcriptional co-activator of various genes such as HAP1 [31 , 32] , FRE2 [33] , ARG1 and ARG4 in cooperation with Gcn4p [34] , TAT1 and TAT2 [35] , and genes induced by Hog1p in cooperation with Sko1p [36] . Genome-wide profiling of changes in nucleosome organization and gene expression that occur following the loss of CYC8 or TUP1 in S . cerevisiae laboratory strains show significant overlap , but additional changes result from the absence of either TUP1 or CYC8 [37] . Thus , the major function of Cyc8p and Tup1p in S . cerevisiae , identified so far , is the repression of pleiotropic gene targets mostly in the form of the Cyc8p-Tup1p co-repressor complex . In addition , several mutually independent repressor functions of Tup1p and Ssn6p ( a functional homologue of S . cerevisiae Cyc8p ) have been reported in Candida albicans , in which filamentous growth and hypha-specific genes are repressed by Tup1p independently of Ssn6p , whereas Ssn6p may act as a repressor of phenotypic switching independently of Tup1p [38] . Ssn6p was recently identified as a negative regulator of the opaque cell transcription program in white C . albicans cells and of the white cell transcription program in opaque cells [39] . Tup1p was reported to be a repressor of the opaque state and , together with its negative regulator Wor1p , has been proposed to control the opaque switch under different circumstances [40] . In addition to its interaction with Tup1p , Ssn6p interaction with histone deacetylase Rpd31p has been reported in C . albicans [41] . In this study , Ssn6p appeared to be a repressor of filamentation as well as of wrinkled colony morphology under particular conditions , independently of Tup1p , and some of these repressive effects were enhanced by deletion of RPD31 . MoTUP1 was recently identified in Magnaporthe oryzae ( a rice pathogen ) , and its deletion causes decreased pathogenicity of the fungus [42] . These studies suggest that Tup1p and Cyc8p play important roles in the pathogenicity of different fungi and that , in addition , these factors could have independent roles . In this study , we provide clear evidence of the functions of the Tup1p and Cyc8p regulators in biofilm colony formation . We present evidence that Cyc8p itself is a repressor of FLO11 gene expression and of the formation of the structured architecture of colony biofilms , whereas Tup1p counteracts Cyc8p , being a positive regulator of FLO11 expression and colony complexity . Furthermore , we show that Tup1p regulates Flo11p accumulation at two different levels—gene expression and Flo11p stability . In addition to Flo11p , other features that are important for colony biofilm formation , such as cell invasiveness , adhesion to solid surfaces and presence of fibers connecting the cells , are also antagonistically regulated by Cyc8p and Tup1p . Conversely , features that are related to other types of multicellularity , such as cell flocculation , are co-repressed by both regulators .
A series of strains was constructed producing different levels of Cyc8p and Tup1p regulators ( Table 1 ) derived from the parental BR-F strain ( wt strain; [43] ) , which forms structured colony biofilms [6] . The tup1 strain ( tup1/tup1 ) was prepared by deleting both alleles of TUP1 , but we did not succeed in preparing a cyc8 strain ( cyc8/cyc8 ) . As the CYC8 gene is essential in the Σ1278 strain-background [44] , resembling in several aspects wild yeast strains , this gene may also be essential in the BR-F strain . Therefore , we constructed strain pGAL-CYC8 ( cyc8/pGAL-CYC8 ) , in which one CYC8 allele is deleted and the second placed under the control of the GAL1-inducible promoter ( pGAL ) , which provides only very low ( basal ) level of CYC8 expression in the absence of galactose . The decreased level of CYC8 mRNA and level of Cyc8p in this pGAL-CYC8 strain ( grown without galactose ) , compared with the BR-F strain , was confirmed by northern blot ( S1A Fig ) and LC-MS/MS ( see below ) , respectively . We also prepared strain pTEF-CYC8 ( CYC8/pTEF-CYC8 ) constitutively over-expressing CYC8 from the TEF1 promoter ( pTEF ) . Unexpectedly , deletion of TUP1 and CYC8 over-expression resulted in a similar , very prominent change in colony architecture indicating opposing roles of Tup1p and Cyc8p in biofilm formation ( Fig 1A ) . In both cases , the strains formed smooth colonies . Conversely , although reduced CYC8 expression slowed the growth of the pGAL-CYC8 strain , this strain formed structured colony biofilms that gradually developed morphology similar to that of wt strain biofilms ( S1B Fig ) . Hence , 3-day-old pGAL-CYC8 colonies exhibited an architecture ( Fig 1B ) with features typical of younger ( 40-h-old ) structured biofilms formed by the wt strain and 5-day-old pGAL-CYC8 colony biofilms resemble 3-day-old biofilms of the wt strain ( S1B Fig and [6] ) . Flo11p is essential for colony biofilm formation . Therefore , we investigated the potential role of Tup1p and Cyc8p in Flo11p expression . We prepared strains pGAL-CYC8-Flo11p-GFP ( cyc8/pGAL-CYC8-Flo11p-GFP ) and pGAL-TUP1-Flo11p-GFP ( tup1/pGAL-TUP1-Flo11p-GFP ) ( derived from the BR-F-Flo11p-GFP strain; [15] ) , in which CYC8 and TUP1 expression is inducible by galactose , to monitor Flo11p-GFP levels in the context of colony biofilm morphology . Presence of galactose in GMA partially affects the colony appearance , slightly reducing the structured morphology , as has similarly been shown for glucose YEPD medium [45] . pGAL-TUP1-Flo11p-GFP and pGAL-CYC8-Flo11p-GFP colonies were , therefore , first grown on GMA plates without galactose for 3 days and then expression of pGAL-controlled genes was induced for ~18 h by adding galactose to wells in the agar . Fig 2A shows that in areas of higher galactose concentration ( near the wells ) , colony morphologies changed due to the induction of TUP1 ( smooth → structured ) or CYC8 ( structured → smooth ) , whereas colonies located far from the galactose source retained their original morphologies . Western blots showed that Flo11p-GFP is produced in high levels in pGAL-TUP1-Flo11p-GFP colonies induced by galactose ( Fig 2B , lane 4 ) , whereas Flo11p-GFP production was totally abolished when CYC8 over-expression was induced by galactose in pGAL-CYC8-Flo11p-GFP colonies ( lane 8 ) . In accordance , Flo11p-GFP was undetectable in pTEF-CYC8-Flo11p-GFP ( CYC8/pTEF-CYC8-Flo11p-GFP ) colonies constitutively overexpressing CYC8 ( lane 5 ) , whereas Flo11p-GFP level in pTEF-TUP1-Flo11p-GFP ( TUP1/pTEF-TUP1-Flo11p-GFP ) ( lane 1 ) colonies was similar to that of wt colonies ( lane 2 ) . Two photon excitation confocal microscopy ( 2PE-CM ) showed that in 3-day-old wt colonies , Flo11p-GFP is present at higher levels in cells at the aerial surface of wt colonies and in cells forming the tips of “roots” invading the agar ( Fig 2C ) . A similar pattern of Flo11p-GFP was observed in structured pGAL-TUP1-Flo11p-GFP colonies near the galactose source and in structured pGAL-CYC8-Flo11p-GFP colonies that were localized far from the galactose source and were thus not induced ( Fig 2D ) . However , Flo11p-GFP was hardly detectable in smooth colonies of both strains . To further clarify regulatory functions of Tup1p and Cyc8p , we compared amounts of TUP1 and CYC8 mRNAs and proteins in wt colonies and colonies of above described strains with differently manipulated levels of Cyc8p or Tup1p ( Fig 3A , left part , and 3B ) . Colonies were grown for 3 days on GMA and then induced by galactose ( or treated with distilled water as a control ) for 4 hours . This induction greatly increased TUP1 and CYC8 mRNA levels , respectively , in pGAL-TUP1 and pGAL-CYC8 strains ( Fig 3A , lanes 4 and 6 , respectively ) . Conversely , amounts of both CYC8 and TUP1 mRNAs were only slightly increased , as compared with mRNA levels in wt colonies , when expression was controlled by the moderate , constitutive pTEF promoter ( lanes 7 and 8 ) . Labeling of both Tup1p and Cyc8p proteins with GFP or 6HA tags resulted in dysfunctional proteins and commercial anti-Tup1p and anti-Cyc8p primary antibodies generated high unspecific background . Therefore , we quantified Tup1p and Cyc8p approximate concentrations in cells from 3-day-old colonies induced/non-induced by galactose for 4 hours by label free LC-MS/MS ( Fig 3B ) . Contrary to TUP1 and CYC8 mRNA levels , which both were highly elevated when expression was induced by galactose ( Fig 3A , lanes 4 and 6 ) , differing enhancement of Cyc8p and Tup1p protein concentration was observed in pGAL-CYC8 and pGAL-TUP1 3-day-old colonies . Whereas Cyc8p level increased only by 40% ( ~1 . 4 times ) , Tup1p level increased more than 5 times as compared with wt colonies ( Fig 3B ) . In accordance with mRNA analysis ( Fig 3A , lines 5 and 3 ) , neither Cyc8p nor Tup1p were detected without galactose induction in pGAL-CYC8 and pGAL-TUP1 colonies , respectively ( Fig 3B ) . Analyses of FLO11 mRNA levels ( Fig 3A ) showed that Cyc8p and Tup1p affect FLO11 gene expression in opposite ways and with differing efficiencies . CYC8 constitutive overexpression in pTEF-CYC8 colonies resulted in absence of FLO11 mRNA ( Fig 3A , lane 7 ) and of Flo11p-GFP protein ( Fig 2B , lane 5 ) , thus confirming Cyc8p as a FLO11 gene repressor . TUP1 deletion in the presence of functional Cyc8p ( tup1 strain ) resulted in the absence of Flo11p ( Fig 2B , lane 9 ) , but a small amount of FLO11 mRNA was still detectable ( Fig 3A , lane 2 ) . This result confirmed that FLO11 transcription is enhanced by Tup1p , but when the TUP1 gene was deleted , some transcription of FLO11 still occurred . In accordance , the level of FLO11 mRNA was significantly reduced after 4 h of galactose induction of pGAL-CYC8 colonies and the level of FLO11 mRNA was significantly increased after 4 h of induction of pGAL-TUP1 colonies ( Fig 3A , lane 6 and 4 , respectively ) . 18 h after galactose induction , Flo11p protein levels increased from non-detectable to a level comparable with the wt strain in pGAL-TUP1 colonies ( Fig 2B , compare lanes 3 and 4; Fig 2D ) and decreased from a wt-like to non-detectable level in pGAL-CYC8 colonies ( Fig 2B , compare lanes 7 and 8; Fig 2D ) . To further examine mutual effects of both regulators , we constructed an additional set of strains derived from the BR-F and BR-F-Flo11p-GFP strains , in which amounts of both regulators were adjustable by the inducing compound ( galactose or copper ) ( Table 1 , last four strains ) . We then evaluated levels of TUP1 , CYC8 and FLO11 gene expression ( mRNA ) and levels of respective proteins . Results of CYC8 and TUP1 mRNA analysis after 4 h of galactose and/or copper induction of colonies of strains pGAL-CYC8/pCUP-TUP1 ( cyc8/pGAL-CYC8/tup1/pCUP-TUP1 ) and pGAL-TUP1/pCUP-CYC8 ( tup1/pGAL-TUP1/cyc8/pCUP-CYC8 ) ( Fig 3A , lanes 9–16 ) were consistent with results of induction experiments performed with strains , in which expression of only one of the regulators was adjustable and the second was controlled by its native promoter ( Fig 3A , lanes 3–6 ) . Only the level of pGAL-regulated mRNA ( of both TUP1 and CYC8 ) was partially diminished when copper was also present during galactose induction ( Fig 3A; compare lanes 14 and 16 for CYC8 and lanes 10 and 12 for TUP1 , decreased level of mRNA is marked by asterisk ) . Since pGAL-regulated expression of CYC8 and TUP1 was also diminished by copper in pGAL-CYC8 and pGAL-TUP1 colonies , respectively ( S2 Fig ) , copper seem to partially reduce transcription from the pGAL promoter . Consistently with pGAL-CYC8 and pGAL-TUP1 induction experiments , increased level of Cyc8p caused a decrease in FLO11 mRNA ( Fig 3A , lanes 11 and 14 ) and in Flo11p concentrations ( Fig 3C , lanes 4 and 7 ) in pGAL-CYC8/pCUP-TUP1 and pGAL-TUP1/pCUP-CYC8 colonies . However , the basal FLO11 mRNA level when neither Cyc8p nor Tup1p was induced ( Fig 3A , lanes 9 and 13 ) was higher than under conditions where a wt-level of Cyc8p was present ( tup1 or pGAL-TUP1 colonies without galactose , Fig 3A , lanes 2 and 3 ) . 4 h-induction of Tup1p by either inducing compound did not significantly increase the FLO11 mRNA level ( Fig 3A , lanes 10 and 15 ) above the basal level identified in the absence of both inducing compounds ( lanes 9 and 13 ) . In fact this basal level was lower in the pCUP-CYC8/pGAL-TUP1 strain than in pGAL-CYC8/pCUP-TUP1 strain ( Fig 3A , compare lanes 9 and 13 ) , possibly because of traces of copper in the medium which can slightly increase CYC8 expression and thus the amount of Cyc8p repressor from the onset of colony growth . Altogether , these data further confirmed that Cyc8p is the main repressor of expression of the FLO11 gene and indicated that Tup1p modulates the level of Cyc8p repressor , potentially via formation of Tup1p-Cyc8p complex ( Fig 4 ) . Analysis of Flo11p-GFP protein levels however suggested an additional function of Tup1p . Flo11p-GFP full length protein was almost undetectable in the absence of both regulators , whereas a high level of free GFP was present in these samples indicating that Flo11p-GFP synthesis was relatively high ( in accordance with high basal level of FLO11 mRNA , Fig 3A , lane 9 and 13 ) , but that the protein was efficiently degraded ( Fig 3C , lanes 2 and 6 , arrows mark free GFP ) . These data indicate dual roles of Tup1p in regulation of Flo11p concentration in colonies and thus in regulation of colony biofilm complexity: counteracting CYC8 repression of FLO11 gene expression and preventing Flo11p degradation , possibly by repressing expression of a specific protease . Flo11p is associated with the cell wall and it is partially shed from cells into the extracellular space [46] . We therefore examined further whether extracellular Flo11p-GFP is degraded and whether presence of Tup1p influences such degradation . As expected , neither Flo11p-GFP nor free GFP was detected in extracellular fluid from colonies of pTEF-CYC8 and tup1 strains ( Fig 3D , lanes 4 and 5 ) , in which FLO11 expression is repressed by Cyc8p . In extracellular fluid from biofilm colonies of wt strain and pGAL-CYC8 strain without galactose , both partially degraded Flo11p-GFP and high level of free GFP were detected ( Fig 3D , lanes 1 and 6 ) , indicating degradation of a fraction of Flo11p-GFP , perhaps during its shedding . In colonies of both pGAL-TUP1/pCUP-CYC8 and pCUP-TUP1/pGAL-CYC8 strains with high basal levels of FLO11 gene expression and protein production in the absence of both inducing compounds , free GFP only was present in extracellular fluid ( Fig 3D , lanes 2 and 3 ) . Consistently , colonies of these strains are smooth . These data indicate that Tup1p prevents degradation of extracellular Flo11p-GFP possibly via repression of expression of a cell wall associated or extracellularly localized protease . Differences in Flo11p processing ( at several positions within the protein ) were found in a strain defective in the kexin Kex2p [46] , serine protease which cleaves precursors of secreted proteins in the trans-Golgi network . However , Flo11p was not identified in the screen of possible Kex2p substrates and does not contain prominent Kex2p cleavage sites ( Lys-Arg at P1 and P2 position ) [47] . Hence , Flo11p is probably not a direct target of Kex2p , but it could be cleaved by another secreted protease , the secretion and/or processing of which requires Kex2p . Next , we examined Cyc8p and Tup1p roles in regulation of other processes that are specific to colony biofilms , such as cell-substrate adhesion and agar penetration and cell-cell interaction via cell wall fibers . Long fibers forming Velcro-like structures in contact sites between the cells were identified in colony biofilms , but not in smooth colonies of the BR-F-flo11 strain , by transmission electron microscopy ( TEM ) of chemically fixed cells [6] . Here we used high-pressure freezing and freeze substitution TEM that improves identification of these structures and revealed some less abundant , extracellular fibrillar material even on the surface of cells within BR-F-flo11 colonies . Fig 5A thus shows that cells in both structured and smooth colonies are covered on their surface by extracellular fibrillar material , which connects adjacent cells . However , the fibers in this material were significantly ( 20–30% ) longer in the structured colony biofilms of the BR-F and non-induced pGAL-CYC8 strains than in the smooth colonies of pTEF-CYC8 , tup1 and flo11 strains ( Fig 5B ) , in which shorter fibers are occasionally visible despite the material appearing to be less structured . The differences are evident also in cell-cell contact sites , where Velcro-like connections were visible among cells in colony biofilms , whereas less structured material was present at contact sites in smooth colonies ( Fig 5C ) . Velcro-like connections may be caused by interaction of N-terminal Flo11A domains of Flo11p as reported in [48] ( Fig 5C , indicated by red mark ) , although direct proof of the presence of Flo11p in these fibers is still lacking . Adhesion to solid surfaces and invasive growth are typical features of fungal biofilms [12] as well as of colony biofilms [13] , which are evident particularly in the area of the colony roots [6] . Cell adhesion and agar invasion are absent in S . cerevisiae flo11 colonies [10 , 13 , 49] . Figs 1 and 6A show that the pTEF-CYC8 and tup1 strains exhibited defects in invasive growth and adhered poorly to the agar . However , similar to the wt strain , cells of the non-induced pGAL-CYC8 strain adhered to the agar even with robust washing . These data show that both organization of extracellular fibrillar material involved in cell-cell contact and cell adhesion to surfaces are antagonistically regulated by Cyc8p and Tup1p . Adhesion to , and invasiveness into agar did not correlate with cell morphology . BR-F colony biofilms were formed by both oval and elongated cells in the aerial part and by pseudohyphae consisting of elongated cells in the subsurface part ( Figs 1B and 6B ) . In contrast , there was a failure to form elongated cells by , not only smooth colony-forming strains with decreased level of Tup1p ( tup1 ) and increased level of Cyc8p ( pTEF-CYC8 ) , but also a colony-biofilm forming strain with reduced level of Cyc8p ( pGAL-CYC8 ) . Thus , although the pGAL-CYC8 strain formed ( in the absence of galactose ) a structured colony biofilm , its root part was formed by chains of rounded cells that divided by monopolar budding and invaded the agar ( Fig 1B ) . Consistently , some wild S . cerevisiae strains form structured colony biofilms despite being unable to form typical pseudohyphae composed of elongated cells [13] . Tup1p and Ssn6p ( Cyc8p ) are repressors of invasive/filamentous growth in C . albicans [38 , 41 , 50] . Our data show that in S . cerevisiae , an imbalance in the Cyc8p and Tup1p levels , rather than the presence or absence of an individual regulator , diminishes cell elongation . This defect , however , does not influence the colony biofilm morphology . The Cyc8p-Tup1p complex has been implicated in the repression of genes involved in cell flocculation , such as FLO1 [17 , 51] . Consistent with the literature [52–54] , either deletion of TUP1 ( tup1 strain ) or substantial reduction of CYC8 expression ( pGAL-CYC8 without galactose ) or reduction of TUP1 and CYC8 expression ( pCUP-CYC8/pGAL-TUP1 or pCUP-TUP1/pGAL-CYC8 strain without inducing compound ) resulted in the formation of macroscopic flocs ( clusters of cells ) that sedimented efficiently ( Fig 6C and 6D ) , indicating de-repression of the flocculation genes . In striking contrast , the wt strain BR-F and the CYC8- or TUP1- over-expressing strains ( pTEF-CYC8 or pTEF-TUP1 strains ) did not form cell clusters . These data show that in contrast to the antagonistic functions of Cyc8p and Tup1p in processes involved in colony biofilm formation , Tup1p and Cyc8p in concert repress other flocculation genes such as FLO1 in the BR-F strain . These results are in agreement with findings showing that i ) in contrast to FLO11 , the expression of other flocculation genes ( FLO1 , FLO9 and FLO10 ) is equivalent in colony biofilms and in smooth domesticated colonies and ii ) citrate buffer treatment and the presence of mannose in the medium ( both of which eliminate the flocculation caused by Flo1p but not that of Flo11p ) affect BR-F cell flocculation in liquid culture but not the adhesiveness of cells from BR-F colonies [43] and iii ) FLO1 and FLO5 play roles in cell aggregation and flocculation , whereas FLO11 expression promotes invasive growth and biofilm formation [55] . Our findings highlight a previously unknown antagonistic function of Tup1p and Cyc8p in the regulation of complexity of yeast colony biofilms . While Tup1p is essential for the formation of colony biofilms , increased levels of Cyc8p prevent formation of colony biofilms leading to formation of smooth colonies similar to those of laboratory strains . The antagonistic functions of Tup1p and Cyc8p are specific to features typical of yeast biofilm life-style , such as cell invasiveness , adhesion to semisolid surfaces and cell-cell adhesion by cell-wall fibers . Properties important for other types of multicellularity , such as cell flocculation [51] , are regulated differently , being repressed by both regulators . In accordance , deletion of genes NRG1 , MIG1 or SFL1 for repressors recruiting the Cyc8p-Tup1p co-repressor complex to promoters [56–58] , did not prevent Cyc8p-mediated repression of colony biofilm formation in pTEF-CYC8 strain ( S3 Fig ) . Flo11p adhesin is key protein in colony biofilm formation affecting most of the above mentioned biofilm-specific processes [10 , 13 , 14] . We therefore tested a hypothesis that Cyc8p and Tup1p regulate biofilm-specific processes via regulation of Flo11p . Indications exist in previous research of a possible relationship between Cyc8p/Tup1p and FLO11 expression , but findings were not consistent . Both positive [59–61] and negative [59] effects of TUP1 deletion on FLO11 mRNA levels have been reported and deletion of the CYC8 gene has been shown to increase FLO11 mRNA levels [56] . Our in depth analyses revealed that Tup1p and Cyc8p regulate the level of Flo11p adhesin in the opposite manner and at different steps in its expression ( Fig 4 ) . Firstly , Cyc8p itself represses FLO11 gene transcription , whereas Tup1p counteracts Cyc8p function , thus contributing positively to FLO11 expression . Efficiency of Cyc8p-based FLO11 gene repression depends on the comparative levels of Cyc8p and Tup1p proteins and we hypothesize that Tup1p can balance the level of free Cyc8p by forming a Cyc8p-Tup1p complex , which apparently does not regulate biofilm specific processes , but can regulate other cellular properties such as expression of flocculins . Four molecules of Tup1p interact with one molecule of Cyc8p [26] , which means that , in accordance with our data , a smaller change in Cyc8p than in Tup1p levels has a stronger effect on FLO11 expression . Secondly , Tup1p also positively regulates the level of Flo11p protein in colony biofilms by preventing its degradation . Flo11p is targeted to the cell wall via the secretory pathway and is partially shed into the extracellular space [46] . The mechanism of its degradation and involvement of specific protease ( s ) are currently unknown . Our data showed that Tup1p prevents degradation of extracellular Flo11p-GFP . Tup1p thus may repress expression of a gene coding for a cell wall protease that is involved in Flo11p degradation . Interestingly , Tup1p represses a set of secreted aspartyl proteinases ( SAPs ) in C . albicans [62] and derepression of genes coding for extracellular proteases was observed in an Aspergilus nidulans strain deleted in the TUPA gene ( an ortholog of TUP1 ) [63] . In summary , our study identifies Cyc8p and Tup1p as important regulators of Flo11p gene expression and protein stability , both affecting the final Flo11p amount in cells and the extracellular space of yeast multicellular structures . According to Flo11p concentration , the structures then acquire different levels of complexity ranging from smooth colonies to colony biofilms . Fine tuning of the amounts and mutual effects of Cyc8p and Tup1p regulators in colony/biofilm cell subpopulations could also provide a mechanism for balancing Flo11p levels and related cellular properties at different positions within the structure and , potentially , at different times during its development . In this respect , further studies are required to uncover potential role of Cyc8p and Tup1p in regulating the amount of Flo11p in different parts of colony biofilms , such as in the internal colony parts , where Flo11p seem not to be present . Orthologs of both Tup1p and Cyc8p are present in different yeast and other fungal species and their role in filamentation , phenotypic switching and virulence of yeast/fungal pathogens has been recently suggested [39–41 , 50 , 64] . Identification of Tup1p as an important positive regulator of the formation of colony biofilms brings practical advantages , making TUP1 orthologs in pathogenic yeasts prospective gene targets for new antifungal treatment strategies .
All strains prepared in this study ( Table 1 ) were derived from the wild yeast strain BR-F from a collection at the Institute of Chemistry ( Slovak Academy of Sciences ) and its derivative BR-F-Flo11p-GFP [15] . Colonies were grown on GMA ( 3% glycerol , 1% yeast extract , and 2% agar ) at 28°C unless otherwise indicated , at densities ranging from 103 to 6 x 103 cells per plate . For the flocculation tests , the strains were grown in liquid GM ( GMA without agar ) . For the strain constructions , G418 and nurseothricin concentrations in GMA were 200 and 100 μg/ml . In galactose/Cu2+ induction experiments , colonies were grown 3 days on GMA . Then , the agar was supplemented by galactose and/or Cu2+ to a final concentrations of either 2% galactose and/or 3 mM CuSO4 for colony incubation for 4 h ( for Northern blot and LC-MS/MS ) or 0 . 1% galactose and/or 0 . 25 mM CuSO4 for longer 18 h incubations used for morphology experiments and determination of Flo11p-GFP levels by Western blots ( lower concentrations of both galactose and Cu2+ were needed to avoid artificially affecting colony morphology in longer incubations ) . Colony images were captured in incident and/or transmitted light . A ProgRes CT3 CMOS camera with a Navitar objective and NIS Elements software ( Laboratory Imaging , s . r . o , Prague , CZ ) were used . Strains with gene deletions and with genes under the control of an artificial promoter ( pCUP , pGAL and pTEF ) were prepared according to [65–67]; primers and plasmids are listed in S1 Table . Yeast cells were transformed as described in [68] . Correct genomic integration of cassettes was verified by PCR using specific primers and by sequencing . Cre-lox system was used to remove antibiotic resistance genes in strains subjected to multiple manipulations [66] . The internal architecture of the microcolonies was visualized by two photon excitation confocal microscopy ( 2PE-CM ) according to [6 , 69] . In brief , colonies were embedded in agarose and cut vertically down the middle . The cut surface was placed on a coverslip , and colony side views were obtained by 2PE-CM . When required , the cross-sections were stained with 1 μg/ml Calcofluor white . Excitation wavelengths of 920 nm and 790 nm and the emission bandwidths of 480–595 nm and 400–550 nm were used for GFP and Calcofluor white . An overview of the morphology of colonies was obtained simultaneously with green GFP fluorescence as autofluorescence in the 600-740-nm wavelength range . Images of the whole colonies ( Figs 1A , 1B , 2C and 2D ) and the central parts of the colonies ( Fig 1B ) were obtained by combining two or three images from neighboring fields of view . The detection of GFP tagged Flo11p ( in cell lysates or the extracellular fluid ) by western blots was performed as described [70] . In brief , cells harvested from colonies were broken by glass beads in the presence of protease inhibitors , and proteins ( 25 μg/lane ) of cell lysates were subjected to SDS-PAGE . GFP was detected by mouse monoclonal horseradish peroxidase ( HRP ) -conjugated anti-GFP antibody ( Santa Cruz ) . Membranes stained by Coomassie blue were used as loading controls ( S4 Fig ) . Extracellular proteins were extracted by phosphate-saline buffer from 3-day-old colonies . After centrifugation , proteins of the supernatant were precipitated by methanol/chloroform treatment [71] . Extracellular proteins extracted from 50 mg of wet biomass were loaded to each slot . For nanoLC-MS-MS analysis , the cells were disrupted in 100 mM triethylammonium bicarbonate buffer using glass beads . Protein aliquots ( 30 μg; determined by the bicinchoninic acid assay , Sigma ) were solubilized using sodium deoxycholate ( 1% ( w/v ) final conc . ) , reduced with tris ( 2-carboxyethyl ) phosphine , alkylated with S-methyl methanethiosulfonate , digested sequentially with trypsin and extracted with ethylacetate saturated with water [72] . Samples were desalted using C18 sorbent ( Supelco pn: 66883-U ) and eluents were dried and resuspended in 20 μl of 1% trifluoroacetic acid . Peptide ( 2 μg ) from each sample were separated on 50-cm C18 column using 2 . 5 h elution gradient and analyzed in a DDA mode on Orbitrap Fusion Tribrid ( Thermo Scientific ) mass spectrometer . Three biological replicates were run for each strain and condition . Resulting raw files were processed in MaxQuant ( v . 1 . 5 . 8 . 3 ) [73] . Searches were performed against latest version of S . cerevisiae Uniprot database and common contaminant database . Further analysis was performed in Perseus ( v . 1 . 5 . 5 . 3 ) [74] . The cell-cell adherence ( flocculation ) assay was performed according to [75] . In brief: 2-day-old cell cultures grown in GM medium were harvested , flocculation disrupted by EDTA ( pH 8 , 50 mM final concentration ) and OD600 of the cell suspension determined ( reading A ) . Then , cells were washed twice by dH2O and suspended in 30 mM CaCl2 . After 60 s , OD600 at upper layers of the cell suspension was measured ( reading B ) . Flocculation ( % ) was calculated according to the formula: 100* ( A-B ) /A . Average of 4 independent measurements +/- SD is shown . The flocs or free cells were photographed using transmission light microscopy ( Microscope DMR , Leica , Germany ) . In the invasive growth assay [76] cells were streaked onto standard GMA plates and grown at 28°C for 3 days . Plates were vigorously washed with water and photographed . Samples were prepared according to [77] with some modifications . Briefly , 3-day- and 5-day-old colonies were frozen in an EM PACT2 high-pressure freezer ( Leica , Germany ) . The samples were freeze-substituted in an automatic FS machine ( Leica , Germany ) in 100% acetone containing 2% osmium tetroxide as follows: -90°C for 96 h , 5°C increase per hour for 14 h , -20°C for 24 h , 3°C increase per hour for 8 h , and 4°C for 18 h . The substituted samples were embedded in pure Epon . Ultrathin sections were cut using a Reichert-Jung Ultracut E ultramicrotome and stained using uranyl acetate and lead citrate . The sections were examined using a JEM-1011 transmission electron microscope ( JEOL , Japan ) operating at 80 kV . Fine structure measurements were performed using a Veleta camera and iTEM 5 . 1 software ( Olympus Soft Imaging Solution GmbH ) . Colonies were suspended in TES buffer ( 10 mM Tris , pH 7 . 5 , 10 mM EDTA , 0 . 5% SDS ) and total RNA isolated using the hot phenol method [78] . Fifteen micrograms of total RNA was denatured in loading buffer with formamide , separated in 1 . 5% agarose gel and transferred to a positively charged nylon membrane ( Amersham Hybond-XL , GE Healthcare Ltd ) . The membranes were hybridized with specific DNA probes prepared using a random primer labeling kit ( Takara ) . The rRNA content was visualized by ethidium bromide staining of gels and used as a loading control ( S2 Fig ) .
|
Yeast biofilms have become an increasingly important clinical problem over the years . Biofilms are often associated with infections resistant to antifungals , which are particularly prevalent in immunosuppressed patients . High resistance is mediated by cell reprogramming leading to the specific organization of internal biofilm structure and development of numerous protective mechanisms including extracellular matrix formation . Colony biofilms , with architecture and protective mechanisms similar to natural biofilms , represent an ideal model for studying molecular mechanisms and regulations behind biofilm development and organization . Here , we describe a new mechanism of antagonistic regulation of biofilm-specific processes and formation of complex colony biofilm structure by Cyc8p and Tup1p transcriptional regulators . Both these regulators are widespread and conserved among yeasts forming clinically important biofilms , including pathogenic yeasts of Candida spp . The identification of Tup1p as a positive regulator of biofilm formation makes Tup1p-orthologs in yeast pathogens potential targets for the design of new strategies of treatment of biofilm infections .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] |
[
"biofilms",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"pathogens",
"microbiology",
"galactose",
"carbohydrates",
"organic",
"compounds",
"regulator",
"genes",
"fungi",
"model",
"organisms",
"animal",
"behavior",
"experimental",
"organism",
"systems",
"gene",
"types",
"zoology",
"animal",
"sociality",
"fungal",
"pathogens",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"mycology",
"behavior",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"chemistry",
"yeast",
"candida",
"eukaryota",
"organic",
"chemistry",
"monosaccharides",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"physical",
"sciences",
"saccharomyces",
"cerevisiae",
"organisms",
"candida",
"albicans"
] |
2018
|
Cyc8p and Tup1p transcription regulators antagonistically regulate Flo11p expression and complexity of yeast colony biofilms
|
AP-2 is the core-organizing element in clathrin-mediated endocytosis . During the formation of clathrin-coated vesicles , clathrin and endocytic accessory proteins interact with AP-2 in a temporally and spatially controlled manner , yet it remains elusive as to how these interactions are regulated . Here , we demonstrate that the endocytic protein NECAP 1 , which binds to the α-ear of AP-2 through a C-terminal WxxF motif , uses an N-terminal PH-like domain to compete with clathrin for access to the AP-2 β2-linker , revealing a means to allow AP-2–mediated coordination of accessory protein recruitment and clathrin polymerization at sites of vesicle formation . Knockdown and functional rescue studies demonstrate that through these interactions , NECAP 1 and AP-2 cooperate to increase the probability of clathrin-coated vesicle formation and to control the number , size , and cargo content of the vesicles . Together , our data demonstrate that NECAP 1 modulates the AP-2 interactome and reveal a new layer of organizational control within the endocytic machinery .
Clathrin-mediated endocytosis is the major vesicular entry route in mammalian cells . The formation of endocytic clathrin-coated vesicles ( CCVs ) depends on a complex protein machinery that deforms the planar plasma membrane into small , cargo-laden vesicles that are released into the cytosol [1] , [2] . The endocytic machinery is organized around AP-2 , a heterotetrameric protein complex in which the N-terminal regions of its two large subunits , α and β2 , together with the two smaller subunits , σ2 and μ2 , form a large globular domain referred to as the trunk [3] . The C-terminal region of each large subunit forms a small , bi-lobed globular domain termed the ear , and the two ears connect to the trunk through flexible linkers . Each of these regions mediates a specific function of the complex , allowing AP-2 to control the recruitment of a myriad of endocytic accessory proteins , cargo , and clathrin to PI ( 4 , 5 ) P2-rich sites of CCV formation at the plasma membrane [1] , [2] . At the protein level , the endocytic machinery is built on the basic principle that short peptide motifs in unstructured or loosely structured regions of one protein mediate low-affinity interactions with a globular domain in a second protein [4]–[6] . In isolation , each interaction is of very low affinity and easily reversible; however , each protein has the potential to simultaneously interact with a number of binding partners to create an interaction network that stabilizes the protein coat around the forming vesicle . Perhaps the least understood step of CCV formation is vesicle initiation , with two main models proposed to explain how new clathrin-coated pits ( CCPs ) are nucleated . In one , the FCHo complex , formed by FCHo1/2 , Eps15 , and intersectin , is recruited to PI ( 4 , 5 ) P2-rich sites at the plasma membrane , marking these sites for subsequent recruitment of clathrin/AP-2 [7] , [8] . In the other , pits initiate by arrival of clathrin/AP-2 to PI ( 4 , 5 ) P2-rich sites with the FCHo complex recruited subsequently [9] . It is likely that in either scenario , clathrin/AP-2 complexes will need to be linked to the FCHo complex; however , the mechanisms that allow for efficient interconnection of the two complexes remain elusive . Each endocytic accessory protein contributes to one or more specific aspects of vesicle formation such as membrane deformation , cargo recruitment , vesicle size control , and vesicle scission; thus , each needs to gain access to vesicle formation sites in the correct temporal order [2] . Many proteins target these sites through interactions with the globular ear domains of AP-2 . During the course of vesicle formation , the β2-ear transitions from accessory protein binding to recruiting clathrin in conjunction with the β2-linker [10]–[12] , whereas the α-ear serves as the main interface for accessory protein recruitment throughout the process [13]–[15] . All α-ear binding partners use one or more of three distinct peptide motifs to bind the ear; DPF/W and FxDxF motifs target the platform subdomain of the α-ear , while WxxF motifs bind the sandwich subdomain [6] , [16] . Yet the mechanisms that modulate α-ear interactions allowing for appropriate recruitment of accessory proteins remain unknown . In many cases , cargo selection and concentration in forming vesicles also depend on AP-2 [1] . Some cargo , such as the transferrin receptor , interact directly with the trunk region of AP-2 . Other cargo , such as the low-density lipoprotein receptor , interact with alternate cargo adaptors , a heterogeneous subclass of accessory proteins that in turn interact with AP-2 and/or clathrin to bridge their cargo to the endocytic machinery [1] , [17] . Therefore , mechanisms that control accessory protein recruitment to forming vesicles also provide a means to modulate vesicle cargo . We identified NECAP 1 and 2 as CCV-enriched proteins using subcellular proteomics [18] , [19] . The two proteins are only 62% identical at the amino acid level , and they function in distinct membrane trafficking processes . Knockdown ( KD ) of NECAP 2 , a ubiquitously expressed isoform , does not appear to influence clathrin-mediated endocytosis but instead inhibits endosomal sorting ( our preliminary data ) . NECAP 1 , which functions in clathrin-mediated endocytosis [19] , is primarily expressed in neurons but is also readily detected in cultured cell lines ( Figure S1 ) , which offer an easy-to-manipulate system to probe for its mechanistic role in this process . Through a WxxF motif at its C-terminus , NECAP 1 binds with high affinity to the α-ear sandwich subdomain of AP-2 [20] . The NECAP 1 N-terminus is well conserved with NECAP 1 orthologs in other species and encodes a globular PH fold , termed PHear [21] . PHear binds to accessory proteins harboring FxDxF motifs , similar to the platform subdomain of the AP-2 α-ear [21] . NECAP 1 binding to the α-ear sandwich site positions PHear in proximity of the α-ear platform , suggesting that NECAP 1/AP-2 complexes act synergistically to control the recruitment of FxDxF-containing accessory proteins to forming vesicles . In the current study we set out to examine the mechanistic role of NECAP 1 in endocytosis and we now report that NECAP 1 works cooperatively with AP-2 to recruit endocytic accessory proteins that control the number , size , and cargo content of CCVs .
The N-terminal region of NECAP 1 ( aa 1–178 ) is highly conserved in NECAP 1 orthologs in other species , and residues 1–133 encode a PH-like domain [21] . We termed the domain PHear , as it has a PH fold and interacts with FxDxF motifs similar to the AP-2 α-ear [21] . Since the conservation extends beyond the C-terminal border of PHear ( aa 129–178 , termed Ex , Figure 1A ) , we reasoned that Ex could also have an important functional role . Ex shows no structural organization by NMR , on its own ( Figure 1B ) or in tandem with PHear [21] . Interestingly , when the entire conserved region of NECAP 1 ( aa 1–178 , termed PHear–Ex , Figure 1A ) was used in affinity-selection assays from rat brain lysate , mass spectrometry of the isolated proteins identified multiple subunits of AP-2 ( unpublished data ) . As the C-terminal WxxF motif that mediates NECAP 1 binding to the α-ear is not located within PHear–Ex ( Figure 1A ) , these data indicate that PHear and/or Ex provide a second mechanism for NECAP 1 to engage AP-2 . Western blot analysis confirmed that endogenous AP-2 is affinity-selected by GST–PHear–Ex ( Figure 1C ) . AP-2 binding is stronger with PHear–Ex than PHear alone ( Figure 1C ) , suggesting the presence of two AP-2 binding sites in the conserved N-terminus of NECAP 1 . This was confirmed using PHear and Ex in isolation , which both bind AP-2 ( Figure 1D ) . The observation that AP-2 binding to PHear–Ex is greater than the combined binding of PHear and Ex in isolation ( Figure 1D ) indicates that the two binding sites cooperate in AP-2 interaction . This is underscored by the fact that Flag-tagged PHear–Ex co-immunoprecipitates endogenous AP-2 from 293-T cell lysates while Flag-tagged PHear alone does not ( Figure 1E ) . Thus , NECAP 1 has three distinct AP-2 binding sites , the high affinity WxxF motif at the C-terminus , and two lower affinity AP-2 binding sites in the conserved N-terminus , one in PHear and one in Ex . To better define the interaction of PHear and Ex with AP-2 , we first mutated K154 and G156 , the only two amino acids within Ex that are invariant throughout evolution ( Figure 1F ) . This double point mutation reduced AP-2 binding of PHear–Ex to levels seen with PHear alone ( Figure 1G ) , further supporting the presence of a site in Ex that interacts with AP-2 . As for PHear , we tested an array of previously generated PHear variants that contain mutations disrupting interactions with FxDxF motifs [21] . With only a few exceptions , mutations that reduce or eliminate binding to FxDxF motif proteins such as amphiphysin I and II also interfere with AP-2 binding ( Figure 1H ) . Therefore , PHear uses an overlapping interface for binding to FxDxF motif proteins and AP-2 , indicating that during CCV formation , NECAP 1 may transition between different roles by changing PHear binding partners . To better understand the multiple PHear–Ex interactions , we first sought to identify the binding site ( s ) for PHear and PHear–Ex in AP-2 . In affinity selection assays , GST-α-ear shows no binding to PHear or PHear–Ex , despite strong binding to full-length NECAP 1 , which contains the C-terminal WxxF motif ( Figure 2A ) . Coupled with the fact that NECAP 1 does not bind the AP-2 β2-ear [19] , it becomes clear that PHear and Ex do not target the canonical AP-2 ear interfaces . We next sought to identify the subunit of the AP-2 heterotetramer involved in PHear–Ex binding . Overlays ( far-Western blots ) provide a powerful means to identify and study protein–protein interactions . For example , synaptojanin was originally identified based on its interaction with the SH3 domains of Grb2 in overlay assays [22] . We thus performed overlay assays on coat proteins stripped from purified CCVs using soluble GST–PHear and GST–PHear–Ex , and both constructs revealed a strong signal at approximately 110 kDa , near the apparent MW of the two large AP-2 subunits α and β2 ( Figure 2B ) . To distinguish between these subunits and further map the site , we immunoprecipitated Flag-tagged deletion variants of α and β2 containing either the trunk and linker or the linker and ear ( Figure 2C ) and used these in overlays . Both PHear and PHear–Ex interact with β2 variants with no binding to the α variants ( Figure 2D ) . We next used variants containing β2 trunk alone , trunk and linker , linker and ear , and ear alone ( Figure 2C ) . PHear and PHear–Ex only bind β2 variants containing the linker ( Figure 2E ) . Throughout these experiments , the presence of Ex did not change the binding pattern of PHear–Ex compared to PHear . Unfortunately , we are unable to map the Ex binding site on AP-2 as Ex in isolation shows limited AP-2 binding ( Figure 1D ) . Mapping the site of PHear interaction to the β2-linker by overlay allowed us to confirm the interaction using in-solution affinity selection assays . GST–β2-linker specifically bound NECAP 1 from brain extracts with binding similar to that seen for clathrin , the only other protein known to interact with the β2-linker ( Figure 2F ) . NMR analysis reveals that the β2-linker binds a site on PHear that overlaps with the interface for FxDxF motifs ( Figure 2G–I ) and titrations of β2-linker with PHear reveal a mean KD of 480 µM ( Figure 2J , K ) . While the affinity of PHear for the β2-linker is low in isolation , full-length NECAP 1 has two additional binding sites on AP-2 that provide avidity effects , the low affinity site in Ex and the high affinity WxxF motif at the C-terminus ( Figure 1C ) . Using C-terminal deletion variants of the β2 trunk and linker in overlay assays , we determined that critical residues for PHear binding are located in the linker between amino acids 633 and 640 ( Figure 2L , M ) . Intriguingly , this area overlaps with the binding motif for the terminal domain of clathrin heavy chain ( Figure 2L ) [23] . Together , these data indicate that PHear binds the β2-linker of AP-2 and that clathrin and NECAP 1 may compete for access to the β2-linker . To test for competition between clathrin and NECAP 1 for AP-2 binding , we performed NMR studies with purified , 15N-labeled PHear in the presence of β2-linker and increasing concentrations of clathrin terminal domain . β2-linker causes chemical shift changes in PHear as it interacts with residues in the PHear binding site and these chemical shift changes are reversed to the ligand-free position in the presence of increasing concentrations of terminal domain ( Figure 3A ) . We next performed more classical in-solution competition experiments . AP-2 was partially purified from stripped CCVs using gel filtration chromatography and bound to GST-clathrin terminal domain immobilized on Sepharose beads . Addition of increasing concentrations of purified PHear–Ex causes a dose-dependent decrease in the binding of AP-2 to clathrin terminal domain ( Figure 3B ) , confirming that PHear–Ex and clathrin compete for binding to the β2-linker . In summary , the data presented in Figures 1–3 indicate that NECAP 1 and AP-2 form two distinct complexes . In the first , the WxxF motif targets the α-ear , while PHear targets the β2-linker . This complex would likely be pre-endocytic because PHear occupies the clathrin-binding site in the β2-linker ( Figure 3C ) . Once vesicle formation is initiated , clathrin competes PHear off the β2-linker , leading to the formation of the second complex , in which NECAP 1 remains bound to the sandwich domain of the AP-2 α-ear through the high-affinity WxxF motif , while PHear and the α-ear platform domain are positioned to cooperate in accessory protein recruitment ( Figure 3C ) . Consistently , we detect cooperation of PHear and α-ear in binding to the FxDxF proteins amphiphysin I and II and synaptojanin 170 ( binding is greater when the two domains are mixed than the sum of binding to the two domains on their own ) ( Figure 3D ) . To address which steps of vesicle formation are dependent on the cooperation of NECAP 1 and AP-2 , we knocked down NECAP 1 in COS-7 cells ( Figure 4A ) , which express NECAP 1 at levels similar to other cultured cell lines ( Figure S1 ) and tested for changes in vesicle formation at the plasma membrane . Two obvious alterations were observed following NECAP 1 KD: ( 1 ) a reduction in the number of AP-2 puncta at the plasma membrane ( Figure 4B , C ) and ( 2 ) an increase in AP-2 signal intensity in a large proportion of the remaining AP-2 puncta ( Figure 4B ) . These AP-2 structures still co-localize with clathrin ( Figure 4D ) , suggesting that they are functional vesicle formation sites . Endocytic vesicles labeled with transferrin following 1 min of uptake are also fewer in number and brighter following NECAP 1 KD ( Figure 4E ) , indicating that the changes occurring at the level of the plasma membrane are maintained in the early endocytic pathway . There is a direct correlation between the immunofluorescence signal intensity of coat proteins such as AP-2 and clathrin and the size of the forming structure [24]–[26] . To examine if the increase in AP-2 signal intensity reflects an increase in the size of vesicle formation sites , we used 3D superresolution microscopy ( Figure 4F–J , Figures S2 and S3 , and Movies S1 , S2 , S3 , S4 ) . Quantification of the diameters of deeply invaginated CCPs in x , y , and z confirms that NECAP 1 KD causes a size increase in all three dimensions ( Figure 4F–H ) . There is also an increase in the number of AP-2 signals detected per vesicle formation site ( Figure 4F , G , I ) , confirming that the increase in AP-2 signal intensity seen by confocal microscopy serves as a reliable readout of increased size . In addition to CCPs , some cell lines form large planar clathrin-coated plaques but only at the membrane opposed to the glass surface ( bottom ) [27] . To address whether NECAP 1 KD leads to an increase in CCP size or to a shift towards coated plaques , we compared the dimensions of AP-2-labeled structures on the membrane contacting the coverslip ( bottom ) and on the cell surface facing the culture medium ( top ) . Both locations show the same three-dimensional increase in diameter ( Figure 4J , Figure S3 ) , indicating that NECAP 1 is needed to control the size of CCPs . To further validate this result , we performed EM analysis ( Figure 4K , L ) . We measured the depth of the pits and placed them into three bins of 0–50 nm , 50–100 nm , and 100+ nm depth . In all three bins , the mean width of the pits was significantly increased in the NECAP 1 KD cells ( Figure 4L ) . Notably , we found no evidence for clusters of CCPs , indicating that NECAP 1 KD causes an increase in pit diameter and not an increased clustering of pits . We next used EM to investigate if the increase in the size of vesicle formation sites translates into larger CCVs . Indeed , the absence of NECAP 1 causes a population-wide shift towards larger CCVs ( Figure 4M ) . The EM data also confirm that despite the increase in size , the clathrin-coated structures still complete the vesicle formation process ( Figure 4M ) . Controlling the size and number of CCVs is critical in all cell systems but nowhere more so than in the presynaptic nerve terminal . Synaptic vesicles are the smallest known transport vesicles and are reformed by clathrin-mediated endocytosis following neurotransmitter release [28]–[32] . Size control is especially important for synaptic vesicles as their size determines the neurotransmitter content [33] . NECAP 1 is expressed at highest levels in brain and is enriched in purified CCVs and synaptic vesicles [19] , [34] . KD of NECAP 1 in primary hippocampal neurons leads to a reduction in synaptic vesicle number and an average increase of 12–14% in synaptic vesicle diameter ( Figure S4 ) , which translates to a change in vesicle volume of nearly 50% . Thus , NECAP 1 serves a similar function in the endocytic machineries of nonneuronal cells and neurons . We assume that the effects on vesicle number and size following NECAP 1 KD are indirect as NECAP 1 functions together with AP-2 to coordinate the recruitment of endocytic accessory proteins during vesicle formation . To determine how NECAP 1 is involved in controlling the number and size of CCVs , we performed rescue experiments with NECAP 1 wild-type and point mutants . One essential feature of NECAP 1 is its ability to interact with the AP-2 α-ear and a NECAP 1 variant in which the high affinity C-terminal WxxF motif is inactivated fails to rescue the NECAP 1 KD phenotypes ( Figure 5A , B ) . In contrast , re-expression of wild-type NECAP 1 restores the number of vesicle formation sites ( Figure 5A , B ) and also leads to lower AP-2 intensity levels in these structures . To address the importance of PHear-mediated interactions , we tested a mutant ( R95E ) that disrupts PHear binding to FxDxF motifs and the AP-2 β2-linker ( Figure 1H ) . This mutant fails to rescue the NECAP 1 KD phenotype ( Figure 5A , B ) despite its ability to target the AP-2 α-ear as seen by its co-immunoprecipitation with AP-2 ( Figure 5C ) . A NECAP 1 mutant in which the AP-2-binding site in Ex was mutated ( K154A/G156S ) ( Figure 1G ) rescues the phenotype ( Figure 5A , B ) . This suggests that the ability of Ex to enhance AP-2 interactions with the NECAP 1 N-terminus is dispensable during vesicle formation . Within full-length NECAP 1 , Ex has by far the lowest affinity to AP-2 and if at all , may only play a role in pre-endocytic NECAP 1/AP-2 interactions . Finally , expression of wild-type NECAP 2 [19] does not rescue the NECAP 1 KD phenotype ( Figure 5A , B ) , demonstrating that the two mammalian NECAP isoforms are functionally divergent . This is consistent with our observations that NECAP 2 is not involved in clathrin-mediated endocytosis but instead functions in endosomal sorting ( our unpublished data ) . Our findings indicate that NECAP 1 functions to modulate the ability of AP-2 to recruit accessory proteins during vesicle formation . We thus set out to determine which accessory proteins might be responsible for the alterations in vesicle size and number seen following NECAP 1 KD . In respect to vesicle size , we tested for an effect of NECAP 1 KD on AP180/CALM . CALM is a clathrin- and AP-2-binding endocytic protein , and the increase in vesicle size in NECAP 1 KD cells is reminiscent of a phenotype observed in CALM KD cells [35] . Similarly , functional disruption of the neuronal isoform of CALM , AP180 leads to an increase in the size of synaptic vesicles [36]–[40] . Both CALM and AP180 contain FxDxF motifs that interact with NECAP 1 [21] , and NMR binding studies demonstrate that an FxDxF-motif peptide derived from AP180 binds to the canonical interface on PHear ( Figure 6A ) . The increase in vesicle size following NECAP 1 KD could thus result from reduced levels of CALM [35] and indeed , in the absence of NECAP 1 , less CALM is observed at sites of vesicle formation ( Figure 6B , C ) . Overexpression of CALM in NECAP 1 KD cells could provide a means to increase CALM at these sites and thus allow the endocytic machinery to rebalance . Indeed , CALM expression in the absence of NECAP 1 led to a decrease in the size of vesicle formation sites as judged by AP-2 intensity ( Figure 6D ) . Thus , NECAP 1 is required for efficient recruitment of CALM to vesicle formation sites where CALM regulates vesicle size , either directly or indirectly . CALM also plays a role in cargo selection by CCPs , even though the cargo-specific effects remain poorly understood . For example , CALM serves as a cargo adapter for R-SNAREs [40]–[42] . In addition , KD of CALM reduces clathrin-dependent endocytosis of amyloid precursor protein and EGF receptor without influencing transferrin receptor endocytosis [43] , [44] . Given that NECAP 1 KD reduces the level of CALM recruited to forming vesicles , we hypothesized that NECAP 1 KD would lead to a selective disruption of cargo entry . Indeed , NECAP 1 KD decreases the clathrin-dependent internalization of EGF by over 40% ( Figure 6E , F ) , similar to the reduction in EGF internalization seen upon CALM KD [44] , while transferrin endocytosis and recycling was not altered ( Figure 6G–I ) . The changes in vesicle size and cargo we observe upon NECAP 1 KD are thus likely a result of the reduced levels of AP180/CALM recruited during vesicle formation . To better understand how NECAP 1 KD causes a decrease in vesicle number , we tested for effects on the FCHo complex . The members of the FCHo protein family form tri-partite complexes with the endocytic accessory proteins Eps15 and intersectin [7] , [8] . FCHo expression levels directly correlate with the number of vesicle formation sites and successful endocytic events , while destabilization of FCHo complexes interferes with vesicle formation [8] , [9] . Interestingly , both FCHo1 and 2 interact with PHear and this binding is reduced in the PHear variant R95A ( Figure 6J ) , which also shows reduced binding to FxDxF proteins and AP-2 ( Figure 1H ) . Testing a range of deletion variants revealed that PHear binds the central linker region of FCHo1 ( Figure 6K ) , which is flanked by the N-terminal FCH domain and the C-terminal mu homology domain . NECAP 1 thus provides a parallel mechanism to Eps15 to interlink the FCHo complex with AP-2 , helping to stabilize and maintain normal numbers of vesicle formation sites . Given the interaction between NECAP 1 and FCHo1/2 , we tested for an effect of NECAP 1 KD on the FCHo complex using FCHo2 as a marker . NECAP 1 deletion causes a decrease in the number of FCHo2 puncta at the plasma membrane ( Figure 6L , M ) , revealing that NECAP 1 is needed to efficiently stabilize the FCHo complex at the membrane . Given the correlation between FCHo expression levels and number of vesicle formation sites [8] , [9] , we reasoned that FCHo overexpression would increase the number of formation sites in NECAP 1 KD cells . Indeed , overexpression of FCHo1 in NECAP 1 KD cells efficiently increased the number of AP-2 puncta at the plasma membrane ( Figure 6N ) . Together , these data demonstrate that the cooperation of NECAP 1 and AP-2 allows for efficient protein recruitment into the endocytic protein network to control important aspects of CCV formation such as vesicle number , cargo content , and size .
Most proteins of the endocytic machinery are categorized as accessory proteins . In general , accessory proteins are recruited to sites of vesicle formation in a temporally controlled manner to facilitate specific steps—for example , the FCHo complex and/or clathrin/AP-2 [7]–[9] initiate vesicle formation at PI ( 4 , 5 ) P2-rich spots on the plasma membrane , epsin and clathrin are involved in membrane deformation during invagination , and a late burst of dynamin recruitment is required for vesicle scission [2] . As such , efficient vesicle formation depends on recruiting the correct set of accessory proteins in sufficient amounts at specific times . However , we have only rudimentary insights into how recruitment of accessory proteins is coordinated and regulated during CCV formation . AP-2 is a central hub of the endocytic machinery , coordinating the recruitment of clathrin , cargo , and numerous accessory proteins . The data presented here identify NECAP 1 as a modulator of AP-2 interactions ( Figure 7 ) . NECAP 1 engages the sandwich domain of the AP-2 α-ear through its C-terminal WxxF motif , while PHear engages the β2-linker at a site overlapping with the binding site for the terminal domain of clathrin . Engagement of the β2-linker by clathrin frees PHear for synergistic interactions along with the platform part of the α-ear for recruitment of endocytic accessory proteins ( Figure 7 ) . Recent studies with endogenously tagged proteins suggest that cargo recognition is not required to stabilize vesicle formation sites [45] . Therefore , the rate and efficiency of vesicle initiation is the major and perhaps only step that controls the endocytic capacity of a cell . However , the low number of molecules involved still hampers our ability to directly study these early events of vesicle formation in great detail . In one model of vesicle initiation , AP-2 stochastically samples the plasma membrane for PI ( 4 , 5 ) P2-rich sites , where it coordinates the recruitment of other endocytic proteins [9] . An alternative model [8] favors the idea that vesicle initiation sites are determined by the recruitment of the FCHo complex to PI ( 4 , 5 ) P2-positive sites at the plasma membrane , with subsequent incorporation of AP-2 and clathrin . If one considers vesicle initiation as a short temporal window of opportunity in which the stochastic association of AP-2 and the FCHo complex with PI ( 4 , 5 ) P2 brings these factors together to allow for their interaction , the success rate of vesicle formation would depend on stabilizing these factors at the membrane long enough to nucleate the formation of the endocytic protein network . NECAP 1 is ideally positioned to accomplish this stabilization , and our data demonstrate that NECAP 1 is required to maintain normal numbers of vesicle formation sites . When the NECAP 1/AP-2 complex is recruited to vesicle initiation sites , the NECAP 1 PHear can release from the β2-linker and cooperate with AP-2 to engage the FCHo complex in a manner parallel to that of Eps15 , thereby increasing the probability of coincidence detection that triggers vesicle formation ( Figure 7 ) . It is tempting to speculate that recruitment of clathrin to initiation sites functions as the switch to trigger vesicle formation , given that the clathrin terminal domain competes PHear off the β2-linker , thereby promoting the cooperation of PHear and α-ear in FxDxF protein binding . Concomitantly , the β2-ear and linker are now fully accessible and can recruit and polymerize clathrin at the site of vesicle formation , providing a scaffold for the growing protein network . The fact that FCHo overexpression in NECAP 1 KD cells overcomes the KD phenotype and rebalances vesicle numbers further attests to the necessity of stabilization factors such as NECAP 1 to concentrate accessory proteins in amounts sufficient for efficient vesicle formation . The composition of the endocytic machinery also controls cargo recruitment . Some cargo is recognized by alternate adaptors , accessory proteins with the ability to link their cargo to AP-2 and clathrin [1] . Interestingly , the endocytic machinery also has an inherent ability to translate cargo content into vesicle size adaptation—for example , vesicles that internalize ligand-bound low-density lipoprotein receptors or viruses increase in size to harbor such large cargo [2] , [24] . On the other hand , synaptic vesicles are recycled such that their characteristic small size is maintained . For neuronal transmission , this is important for two reasons: first , smaller vesicles can be formed faster and this might be needed to maintain synaptic vesicle pools during times of high neuronal activity . Second , the size of a vesicle directly determines vesicle volume , which in turn determines the quantal amount of neurotransmitter , with implications for the strength of synaptic transmission and synaptic potentiation [46] . Consistently , alterations in synaptic vesicle size have been demonstrated to cause changes in neurotransmission in nonmammalian and mammalian systems alike [39] , [44] , [46] , [47] . The predominantly neuronal expression pattern of NECAP 1 , together with the fact that NECAP 2 does not function in clathrin-mediated endocytosis , suggests that a tightly controlled and efficient recruitment of accessory proteins may be most critical at the synapse . However , the molecular mechanisms that determine vesicle size and couple cargo to size regulation remain elusive . In nonneuronal cells , deletion of CALM causes an increase in vesicle size [35] . Similarly , deletion or mutation of the neuronal isoform AP180 leads to larger synaptic vesicles [36]–[40] , demonstrating the importance of these accessory proteins for vesicle size control even though the precise mechanism of their function remains unclear . Both proteins also have cargo-specific effects , AP180 and CALM recruit synaptobrevin during synaptic vesicle recycling , and CALM also promotes EGF receptor and amyloid precursor protein internalization in nonneuronal cells [38] , [40] , [43] , [44] , [47] . Our study reveals that the cooperation of AP-2 and NECAP 1 is crucial for efficient CALM recruitment and suggests that decreased CALM levels due to NECAP 1 ablation cause the formation of larger vesicles . This is confirmed by the fact that CALM overexpression in NECAP 1 KD reverses the phenotype and allows for the formation of small vesicle formation sites in the absence of NECAP 1 . It remains formally possible that the size increase in vesicle formation sites observed following NECAP 1 KD results from clustering of vesicle formation sites or an increase in endocytic hot spots [48] , but our EM analysis of vesicle formation sites does not support these possibilities . Live-cell imaging reveals virtually identical recruitment profiles of CALM and NECAP 1 during vesicle formation [49] , strongly supporting the idea that NECAP 1 and CALM share a common function . Moreover , NECAP 1 KD specifically reduces clathrin-dependent endocytosis of EGF receptor with no effect on transferrin receptor internalization , reminiscent of the phenotypes specific for CALM KD [44] . Together , these data demonstrate the importance of NECAP 1 for efficient accessory protein recruitment to sites of vesicle formation and reveal how balancing the levels of endocytic proteins during vesicle formation in turn controls vesicle size and cargo selection .
Mouse monoclonal antibodies against CHC ( clone 23 ) , α-adaptin ( clone 8 , for Western blotting ) , γ-adaptin ( clone 88 ) , and EEA1 ( clone 14 ) were from BD Transduction Laboratories ( Lexington , KY ) . Mouse monoclonals against α-adaptin ( AP . 6 , for immunofluorescence ) and Flag ( M2 ) were from Thermo Scientific and Sigma ( St-Louis , MO ) , respectively . Rabbit polyclonal antibody against c-myc ( A-14 ) and goat polyclonal antibody against CALM ( C-18 ) were from Santa Cruz and polyclonal antibodies against amphiphysin I/II ( 1874 ) , clathrin light chains , and NECAP 1 and 2 have been described previously [19] , [20] , [50] , [51] . The antibody against synaptojanin 170 was a generous gift of Dr . Pietro De Camilli ( Yale University ) . AlexaFluor 633–conjugated human transferrin ( T-23362 ) and biotinylated EGF complexed to Texas Red-Streptavidin ( E-3480 ) were from Invitrogen and Cy3-conjugated human transferrin ( 009-160-050 ) was from Jackson ImmunoResearch ( West Grove , PA ) . The mCherry tag was detected by Western blot using a mouse monoclonal antibody against RFP ( abcam , Cambridge , MA , ab65856 ) . The synthetic amphiphysin I peptide was purchased from the HHMI/Keck Biotechnology Resource Laboratory , Yale University . The synthetic peptides for AP180 and the β2-linker were purchased from Sheldon Biotechnology Center at McGill University . The following constructs were described previously: GST–NECAP 1 , GST–α-ear , and Flag–PHear–Ex ( termed aa1–178 ) [19] , Flag–NECAP 1 and Flag–NECAP 1 W272A , F275A ( termed AVQA ) [15] , GST–PHear , GST–PHear–Ex ( termed aa1–178 ) , GST–Ex ( termed aa129–178 ) , and the GST–PHear point mutants [21] . cDNA clones for NECAP 1 ( gi:27229051 ) , NECAP 2 ( gi:13384759 ) , α-adaptin ( gi:163644276 ) , β2-adaptin ( gi:71773037 ) , FCHo1 ( gi:255683302 ) , and FCHo2 ( gi:30854355 ) were used as PCR templates , and point mutations as needed were introduced using the megaprimer procedure [52] . For bacterial expression , inserts were subcloned into pGEX-4T1 or pGEX-6P1 . For mammalian expression , inserts were subcloned into pcDNA3 with integrated Flag- or myc-tags ( described in [20] , [53] ) . For NMR studies using purified Ex , a PCR-amplified DNA fragment encoding amino acids 128–178 was subcloned into pPROEX-HTb for the expression of N-terminally His-tagged protein . For NECAP 1 KD , NECAP 1–specific target sequences for human and rat protein were designed using the Block-iT RNAi Designer ( Invitrogen ) and subcloned into pcDNA6 . 2/GW-EmGFP-miR ( Invitrogen ) following the manufacturer's instructions . The EmGFP-miR cassette was then amplified by PCR and subcloned into the pRRLsinPPT vector to generate the microRNA expression vectors . The number given in the name of each KD virus corresponds to the first nucleotide position targeted in the mRNA . The control virus has been described previously [54] . For generation of rescue/protein expression viruses , the microRNA part of the EmGFP expression cassette in pRRLsinPPT was replaced by a polylinker , which was subsequently used to clone Flag-tagged variants of NECAP 1 and 2 in frame with EmGFP such that EmGFP–Flag–NECAP fusion proteins were expressed . Expression vectors for mCherry-tagged FCHo1 and 2 were from addgene ( Cambridge , MA , plasmids 27690 and 27686 , respectively ) . The construct for GST-clathrin terminal domain was a gift of Dr . James Keen ( Thomas Jefferson University , Philadelphia , PA ) . HEK-293-T and COS-7 cells were maintained in DMEM High Glucose ( Invitrogen ) containing 10% FBS ( PAA Laboratories Inc . ) , 100 U/ml penicillin , and 100 µg/ml streptomycin ( both Invitrogen ) . For expression of VSVG pseudotyped virus , HEK-293-T were seeded with 107 cells/plate on 15 cm plates in 25 ml of regular culture medium , with six plates for each virus . The following day , each microRNA or protein expression vector was co-transfected with a packaging mix ( containing pMD2 . g , pRSV-Rev , and pMDLg/pRRE , Addgene ) using calcium phosphate . After 8 h , the medium was removed from each plate and replaced with 15 ml of collection medium per plate ( regular medium supplemented with 1× nonessential amino acids ( Gibco ) and 1 mM sodium pyruvate ( Gibco ) ) . At 24 , 36 , and 48 h posttransfection , the medium was removed from each plate and for the 24 and 36 h time point , replaced with 15 ml collection medium . The supernatants for each construct and each collection were combined and stored at 4°C until the end of the collection procedure . The supernatants were then filtered through a 0 . 45 µm PES membrane and the virus was concentrated by centrifugation ( 8 h at 17 , 000× g ) , and the resulting pellets were resuspended in DMEM in 1/2 , 000 of the original volume . Concentrated virus was stored at −80°C until use . To determine the virus titer , HEK-293–T cells were plated in 24-well plates with 40 , 000 cells/well in regular medium . At 10–14 h postplating , the medium was replaced with regular medium containing varying amounts of concentrated virus . The next day , 1 ml regular medium was added to each well . Three days after transduction , the medium was replaced with 300 µl PBS to allow better visualization of the GFP fluorescence , and transduction efficiency was calculated based on the percentage of GFP-positive cells for the different virus dilutions . The MOIs used for COS-7 cells and primary hippocampal neurons are arbitrary MOIs based on the HEK-293–T cell transduction rate . Statistical tests and posttests used are indicated in the figure legends where appropriate . N indicates the number of independent repeats analyzed , and n indicates the size of the total pool analyzed , if applicable . For KD studies in COS-7 cells , cells were plated on the day of transduction . For transduction , the culture medium was replaced by DMEM High Glucose ( Invitrogen ) supplemented with 2% heat-inactivated FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , and 6 µg/ml polybrene ( Sigma ) , and viruses were added at an MOI of 10 . The next day , media was replaced with fresh culture medium and the cells were cultivated until assays were performed 6 d after transduction . In some cases , COS-7 KD cells were transfected using jetPRIME ( Polyplus Tranfection ) following the manufacturer's instruction 5 d after transduction and processed for immunofluorescence following overnight incubation . For rescue studies , KD cells were plated on coverslips on day 5 after transductions . On the same day , cells were transduced with rescue viruses using an MOI of 4 as described above . The media was replaced early the next morning , and cells were processed for immunofluorescence 24 h after the second transduction . For analysis of KD COS-7 cells , cells plated on poly-L-lysine-coated coverslips were processed for immunofluorescence following standard protocols 6 d after transduction with for COS-7 cells . COS-7 cells transduced with protein expression viruses or transfected with expression constructs were processed 1 or 2 d after manipulation . Images were analyzed using Image J ( National Institutes of Health , Bethesda , MD ) . Control and NECAP 1 KD cells ( nt220 ) were plated with 40 , 000 cells/well on poly-L-lysine-coated 8-well Lab-Tek II chambered coverglasses #1 . 5 ( cat . no . 155409 ) . The following day , cells were fixed with 2% paraformaldehyde for 10 min at RT , processed for immunofluorescence detection of endogenous AP-2 using Alexa647-conjugated secondary antibodies , and stored in PBS until imaging . Images were recorded with a SR 200 microscope ( Vutara , Inc . , Salt Lake City , UT ) based on the Biplane FPALM approach [55] . The system features four laser lines ( 405 , 488 , 561 , and 647 nm ) for excitation and activation of single fluorescent molecules . Speckle-free illumination with an even intensity distribution is realized by a specialized beam homogenizer . Images of fluorescing molecules are recorded with a 60×/1 . 2NA Olympus water immersion objective on an Photometrics Evolve 512 EM-CCD camera with the gain set at 50 . Each acquisition consisted of 7 , 000 frames recorded at a speed of 40 frames/s , which encompassed a 20×20 µm field of view . The maximum powers used for the readout laser ( 647 nm ) and activation laser ( 405 nm ) were 4 and 0 . 05 kW/cm2 , respectively . The calibration entails experimentally calculating the point spread function ( PSF ) in three dimensions . This was done using 100 nm Tetraspeck beads . The analysis and rendering were done using Vutara's SRX localization and visualization software , based on an enhanced implementation of Juette et al . [55] and Mlodzianoski et al . [56] . Data were analyzed by the Vutara SRX software ( version 3 . 16 ) . In short , particles were identified by their brightness from the combined images taken in both planes simultaneously . If a particle was identified in multiple subsequent camera frames , data from these frames were combined for the specific identified particle . Background can be optionally removed based on the observed signal in the frames before or after the frames in which a particle was observed . Identified particles were then localized in three dimensions by fitting the raw data in a customizable region of interest ( typically 16×16 pixels ) centered around each particle in each plane with a 3D model function that was obtained from recorded bead datasets . The recorded fields are aligned automatically by computing the affine transformation between the pair of planes . Sample drift can be corrected by cross-correlation of the determined localized particles [56] or tracking of fiduciary markers . Fit results were stored as data lists for further analyses . The SRX software allows the 3D display of localized particles as solid shaded spheres or as an accumulation of transparent gaussian kernels . Alternatively , 2D slices through the 3D volume in any of the three main directions can be shown . COS-7 cells transduced with control and NECAP 1 KD viruses were starved in DMEM High Glucose overnight . For microscopic analysis , cells were chilled on ice for 30 min and then incubated with Cy3-transferrin ( 200 µg/ml ) in ice-cold DMEM on ice for 1 h . Cells were washed with cold PBS and incubated in prewarmed culture medium at 37°C for the times indicated . At each time point , a sample of cells was chilled on ice , surface-bound transferrin was removed by acid wash ( 0 . 2 M acetic acid , 0 . 5 M NaCl ) , followed by a PBS wash . The cells were fixed with 4% PFA and processed for immunofluorescence detection of marker proteins if indicated . For FACS analysis , cells were chilled on ice for 30 min and then incubated with Alexa633-transferrin ( 200 µg/ml ) in ice-cold DMEM on ice for 1 h . Cells were washed with cold PBS and incubated in prewarmed culture medium at 37°C for the times indicated . At each time point , a sample of cells was chilled on ice , surface-bound transferrin was removed by acid wash ( 0 . 2 M acetic acid , 0 . 5 M NaCl ) , followed by a PBS wash . The cells were removed from the plate in 1 ml of PBS by pipetting , filtered through a cell strainer , and analyzed by flow cytometry on a FACSCalibur ( Becton Dickinson ) . For EGF internalization assays , control and NECAP 1 KD cells were maintained in regular culture medium at 37°C . The regular medium was replaced with prewarmed DMEM High Glucose medium containing 2 ng/ml of Texas Red-labeled EGF for 2 . 5 min at 37°C . The cells were then chilled on ice and surface-bound EGF was removed by acid wash , followed by a PBS wash . The cells were fixed with 4% PFA , and internalized Texas Red-EGF was detected by fluorescence microscopy . For comparison , parallel sets of cells were processed to detect Cy3-transferrin ( 200 µg/ml ) internalization under these condition . COS-7 cell monolayers were washed in 0 . 1 M sodium cacodylate buffer ( Electron Microscopy Sciences ) and fixed in 2 . 5% glutaraldehyde ( Electron Microscopy Sciences ) in sodium cacodylate buffer overnight at 4°C . The following day the cells were washed in 0 . 1 M sodium cacodylate buffer and incubated in 1% osmium tetroxide ( Mecalab ) for 1 h at 4°C . The cells were dehydrated in a graded series of ethanol/deionized water solutions from 50%–100% . The cells were then infiltrated with a 1∶1 and 3∶1 Epon 812 ( Mecalab ) ∶ethanol mixture , each for 30 min , followed by 100% Epon 812 for 1 h for embedding in the wells , and polymerized overnight in an oven at 60°C . The polymerized blocks were trimmed and 100 nm ultrathin sections cut with an UltraCut E ultramicrotome ( Reichert Jung ) and transferred onto 200-mesh copper grids ( Electron Microscopy Sciences ) having formvar support film . Sections were poststained for 8 min with 4% uranyl acetate ( Electron Microscopy Sciences ) and 5 min with lead citrate ( Fisher Scientific ) . Samples were imaged with a FEI Tecnai 12 transmission electron microscope ( FEI Company ) operating at an accelerating voltage of 120 kV and equipped with a Gatan 792 Bioscan 1k61k Wide Angle Multiscan CCD Camera ( Gatan , Inc . ) . Vesicle and pit size was measured using Image J ( National Institutes of Health , Bethesda , MD ) . PHear GST fusion protein and Ex His-tag fusion protein were expressed in the E . coli strain BL21 . To generate uniformly 15N-labeled protein , the cells were grown in M9 minimal media containing 15NH4Cl . Bacteria were induced at 30°C for 4 h using IPTG at a final concentration of 1 mM once OD at 600 nm reached 0 . 8 . The GST-fusion protein was purified , cleaved with thrombin in PBS , and thrombin was removed using benzamidine-Sepharose . The protein samples were further purified by S75 gel filtration equilibrated with PBS . The Ex His-tag fusion protein was purified by Ni-charged chelating Sepharose in 8 M guanidium chloride in PBS and eluted in 2 M guanidinium chloride in PBS with 500 mM imidazol . The sample was then purified by gel filtration using a S75 column equilibrated with PBS , immediately cleaved by TEV protease , then purified by reverse phase HPLC using a C18 column . The purified protein was lyophilized to afford a white powder . The NMR samples contain freshly prepared buffer with 25 mM sodium phosphate pH 7 . 2 , 75 mM NaCl , 0 . 5 mM EDTA , and 3 mM DTT . All NMR experiments were performed at 30°C using a Bruker DRX 600-MHz spectrometer . Spectra were processed by NMRPipe [57] and analyzed by NMRview [58] . NMR titrations were carried out by acquiring 1H-15N heteronuclear single quantum correlation ( HSQC ) spectra on 250 µL of 15N-labeled protein at a concentration of 0 . 1–0 . 2 mM . Subsequent spectra were taken after the addition of an unlabeled ligand . Analysis of peptide binding to the PHear domain was carried out by comparison of chemical shifts for backbone amide signals in 15N–1H HSQC spectra . Weighted average shifts ( ( Δδ15N ) 2+ ( Δδ1H ) 2 ) 0 . 5 were used to identify binding site residues . The NMR assignments for the NECAP 1 PHear domain were previously determined [16] . Peptides for AP-180 ( Ac-VDIFGDAFAAS ) and the β2-linker ( Ac-SQGDLLGDLLNLDLPPVN ) were purified by reverse phase C18 HPLC , lyophilized , and dissolved in NMR buffer to make peptide concentrations of 4 . 1 mg/mL and 5 . 1 mg/mL , respectively . 15N-labeled PHear at a concentration of 0 . 15 mM was titrated with unlabeled peptides at relative concentrations of 2∶1 , 1∶1 , 1∶2 , 1∶4 , 1∶8 , and 1∶16 . To observe that the PHear domain competes with clathrin for access to the β2-linker , 15N-labeled PHear together with unlabeled β2-linker peptide were titrated with unlabeled clathrin terminal domain ( aa 1–579 ) . The relative concentrations of PHear , β2-linker , and clathrin terminal domain were 1∶0∶0 , 1∶4∶0 , 2∶8∶1 , 1∶4∶1 , and 1∶4∶2 , respectively . CCVs were purified from adult rat brain using buffer A ( 100 mM MES , pH 6 . 5 , containing 1 mM EGTA , and 5 mM MgCl2 ) as described previously [59] . For the extraction of coat proteins , CCVs were centrifuged at 200 , 000× g for 15 min , the pellet was resuspended in 0 . 5 M Tris pH 7 . 0 , 2 mM EDTA , and incubated for 30 min on ice . The samples were centrifuged at 200 , 000× g and the supernatant fraction was resolved by SDS-PAGE , transferred to nitrocellulose and processed for Western blotting or overlay assays . For affinity selection assays , soluble cell and rat brain extracts for affinity selection assays were prepared in 10 mM HEPES , pH 7 . 4 , 1% Triton X-100 , 50 mM NaCl , 0 . 83 mM benzamidine , 0 . 23 mM phenylmethylsulphonyl fluoride , 0 . 5 µg/ml aprotinin , and 0 . 5 µg/ml leupeptin and incubated for 1 h at 4°C with GST fusion proteins precoupled to glutathione-Sepharose . For AP-2 co-immunoprecipitation assays from 293-T cell lysates transfected with Flag-tagged NECAP 1 variants , the NaCl concentration in the buffer was reduced to 33 mM and lysates were incubated with protein G-agarose alone ( mock ) or with protein G-agarose and 10 µg Flag-antibody ( M2 ) for 1 h at 4°C . Co-immunoprecipitation studies of endogenous AP-2 from brain cytosol and solubilized membrane fraction were performed as described previously [19] . Proteins were resolved by SDS-PAGE and analyzed by Western blotting . Overlays ( far-Western blots ) were performed as described previously [60] . CCVs were purified from rabbit brain ( Pel-Freez , 250 g ) and coat proteins were prepared for gel filtration as previously described [61] . Coat proteins were separated on a Hiprep 26/60 Sephacryl S-300 HR size exclusion column and eluted at 0 . 5 ml/min . Fractions of 1 ml were collected , AP-2-containing fractions identified by Western blot , and stored at −80°C until use . For competition assays , one 1 ml fraction was precipitated with ammonium sulfate , spun for 30 min at 10 , 000× g , and the resulting pellet dissolved in binding buffer ( 10 mM HEPES , pH 7 . 4 , 1% Triton X-100 , 50 mM NaCl , 0 . 83 mM benzamidine , 0 . 23 mM phenylmethylsulphonyl fluoride , 0 . 5 µg/ml aprotinin , and 0 . 5 µg/ml leupeptin ) . Equal aliquots of the purified AP-2 solution were incubated for 1 h at 4°C with 40 µg of purified GST-clathrin terminal domain or GST alone precoupled to glutathione beads . Unbound proteins were removed by washing three times with 1 ml of binding buffer . Equal fractions of GST-clathrin terminal domain were incubated 1 h at 4°C with either 1 ml of binding buffer alone or with 1 ml of binding buffer supplemented with varying amounts of purified PHear–Ex . Unbound proteins were removed as before and the samples were resolved by SDS-PAGE , and AP-2 binding to the clathrin terminal domain was analyzed by Western blot .
|
Clathrin-mediated endocytosis is the major entry portal for cargo molecules such as nutrient and signaling receptors in eukaryotic cells . Generation of clathrin-coated vesicles involves a complex protein machinery that both deforms the membrane to generate a vesicle and selects appropriate cargo . The endocytic machinery is formed around the core endocytic adapter protein AP-2 , which recruits clathrin and numerous accessory proteins to the site of vesicle formation in a temporally and spatially controlled manner . Yet it remains elusive as to how these interactions are regulated to ensure efficient vesicle formation . Here we identify the endocytic protein NECAP 1 as a modulator of AP-2 interactions . We show that NECAP 1 and AP-2 form two functionally distinct complexes . In the first , NECAP 1 binds to two sites on AP-2 in such a manner as to limit accessory protein binding to AP-2 . Recruitment of clathrin to vesicle formation sites displaces NECAP 1 from one of these sites , leading to the formation of a second complex in which NECAP 1 and AP-2 cooperate for efficient accessory protein recruitment . Through these interactions , NECAP 1 fine-tunes AP-2 function and the two proteins cooperate to increase the probability that a vesicle will form and to determine the size and cargo content of the resulting vesicle .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
NECAP 1 Regulates AP-2 Interactions to Control Vesicle Size, Number, and Cargo During Clathrin-Mediated Endocytosis
|
Simian Virus 40 Large Tumor Antigen ( LTag ) is an efficient helicase motor that unwinds and translocates DNA . The DNA unwinding and translocation of LTag is powered by ATP binding and hydrolysis at the nucleotide pocket between two adjacent subunits of an LTag hexamer . Based on the set of high-resolution hexameric structures of LTag helicase in different nucleotide binding states , we simulated a conformational transition pathway of the ATP binding process using the targeted molecular dynamics method and calculated the corresponding energy profile using the linear response approximation ( LRA ) version of the semi-macroscopic Protein Dipoles Langevin Dipoles method ( PDLD/S ) . The simulation results suggest a three-step process for the ATP binding from the initial interaction to the final tight binding at the nucleotide pocket , in which ATP is eventually “locked” by three pairs of charge-charge interactions across the pocket . Such a “cross-locking” ATP binding process is similar to the binding zipper model reported for the F1-ATPase hexameric motor . The simulation also shows a transition mechanism of Mg2+ coordination to form the Mg-ATP complex during ATP binding , which is accompanied by the large conformational changes of LTag . This simulation study of the ATP binding process to an LTag and the accompanying conformational changes in the context of a hexamer leads to a refined cooperative iris model that has been proposed previously .
Helicases are a family of ATPase motors that couple the energy of ATP binding and hydrolysis to conformation changes , which in turn is coupled to the unwinding and translocation of DNA [1] . Simian Virus 40 ( SV40 ) large tumor antigen ( LTag ) is an efficient hexameric helicase that belongs to the helicase superfamily III , as well as the AAA+ protein family . The high resolution structures of LTag hexameric helicase in different nucleotide binding states have been previously reported [2] , [3] , including the Apo , the ATP-bound and the ADP-bound states . These three structures reveal an iris-like motion of the hexamer helicase during the drastic conformational switches that are triggered by ATP binding and hydrolysis . Accompanying the iris-motion of the LTag hexamer is the longitudinal movements of the six β-hairpins along the central channel . Despite the advancement in LTag helicase studies mentioned above , the detailed paths for these conformational switches and the corresponding energetics associated with the ATP binding process are unknown , which can be simulated by a computational approach using molecular dynamics and targeted molecular dynamics . Molecular dynamics ( MD ) propagates the molecular system under the laws of classic mechanics [4] , [5] , and is suitable for studying conformational changes . However , the current computational capability restricts the size ( molecular weight ) of the studied system and the time scale of MD simulation . For the studies of larger and more complex systems , targeted molecular dynamics ( TMD ) has been used to accelerate the simulation , which adopts an additional holonomic constraint on the physical potential to reduce the root mean square deviation between the current structure and the final ( targeted ) structure [6] . TMD is suitable to calculate the transition pathways between two known protein conformations . The combination of MD and TMD methods have been widely applied to the dynamics studies of various systems , such as the Ras p21 in the signal transition pathway [7] , F1-ATPase system [8] , [9] , the GroEL complex [10] , and the human a-7 nAChR receptor [11] . Here we adopted TMD to calculate the whole transition pathway and used MD to simulate the accurate conformational change in certain key time slots . In order to understand the energetic aspects of ATP triggered conformational changes of LTag hexameric helicase , we simulated the ATP binding process of LTag and the associated conformational changes . We first built an Apo state with six ATPs placed 20 Å away from the binding pocket of the original Apo structure . Then we used the TMD approach to calculate the transition pathway from the Apo state to the ATP bound state and examined the ATP binding process that powered this conformational transition . The results suggest an ATP molecule goes through a three-step process before being “locked” inside the nucleotide pocket . Meanwhile , the configurations of the binding pocket along the ATP binding pathway were evaluated by using the linear response approximation ( LRA ) version of the semi-macroscopic protein dipoles langevin dipoles method ( PDLD/S ) , a method that is capable of evaluating binding free energies significantly faster than the microscopic methods with comparable accuracy [12] . In addition , the simulation results of the conformational transition reveal a refined pathway for the cooperative iris-like movement of LTag hexamer helicase previously observed in crystal structures .
There are three high-resolution LTag hexameric helicase structures corresponding to different nucleotide bound states [2] , namely , the Apo state ( PDB ID 1svo ) , the ATP bound state ( PDB ID 1svm ) and ADP bound state ( PDB ID 1svl ) . The hexameric helicase structure reveals two stacked hexamer rings with a central channel ( Fig . 1 ( A ) ) . Each LTag subunit of the hexamer contains three structural domains , D1 , D2 and D3 ( Fig . 1 ( B ) ) . D1 is the N terminal Zn domain essential for LTag hexamerization [2] , [13] . D2 is a typical AAA+ domain with Walker A or p-loop and Walker B motifs , which is important for ATP binding [2] . D3 is composed mostly of long helices , which is sequentially interrupted by D2 roughly in the middle of D3 , while D1 is structurally well separated from D2/D3 ( Fig . 1 ( C ) ) . The binding pocket is located at the interface between two adjacent monomers . The monomer with the P-loop at a given interface is named cis-monomer , and the other monomer forming the interface is named trans-monomer . For ATP to bind to the nucleotide pocket , the only possible route is through the opening between the two neighboring monomers ( or subunits ) from the C-terminal end ( bottom ) . The binding pocket residues on the cis-monomers can be divided into two groups , the I428 , D429 , K432 , T433 , T434 on the P-loop , and the N529 , D474 on the Sensor I motif ( Fig . 1 ( C ) ) . The binding pocket residues on the trans-monomers include the arginine finger tR540 ( t designates trans ) and lysine finger tK418 , and residues tR540 , tD502 and tR498 . The ATP interacts with the cis-residues and trans-residues mainly through its phosphate group and the ribose . The adenosine group inserts into the hydrophobic pocket formed between two helices , h9 and h13 , on the cis-monomer ( Fig . 1 ( C ) ) . There are three major conformational transition stages of the LTag molecular motor , which is associated with the ATP binding stage , followed by the ATP hydrolysis and the ADP releasing stages . In this report , we focus on the study of the ATP binding stage . We have built a pre-Apo state model by placing six ATP molecules 20 Å away from the nucleotide binding pocket , and an ATP docking stage model by putting the ATP in the binding pocket of the Apo state . We simulated a 1 ns ( nanosecond ) pathway from pre-Apo state to the Apo ATP binding states and a 1 ns pathway from pre-Apo to the ATP docking state and finally to the ATP binding state . To study the LTag helicase overall conformational changes during ATP binding , we took a closer look at the ATP binding pocket by analyzing the trajectory involving the ATP ligand binding process . In this section , we studied the conformational changes of the cis and trans residues involved in ATP binding , the movement of ATP , and the dynamic hydrogen bond formation during the ATP binding process . The results of this study suggest a cross-locking model of the binding pocket for ATP binding . The coordination of Mg2+ plays an important role in the ATP binding . Similar experimental study of F0F1-ATPase shows that the addition of Mg2+ will increase the binding affinity of the nucleotide and helps to proceed to the tight binding state [8] , [17] . The binding pocket residues coordinate with the Mg2+ ion directly or through some intervening water molecules . Among these intervening water molecules , the apical water , WAT1 , near the γPi helps to stabilize the pocket residues , and attack the γPi during hydrolysis . In our simulation , we observed the coordination of Mg2+ with the intervening waters during the ATP binding procedure ( Fig . 6 ) . The Mg2+ has a strong propensity to assume an octahedral coordination [17] . During the ATP docking stage , the Mg2+ ion forms a complex of six-element structure with the β , γ oxygen and four water molecules . The whole complex ( ATP-Mg2+ and five coordinated water molecules ) docks into the binding pocket until the WS . There is a flattening stage in the distance profile between cis-residue T433 and the Mg2+ ( Fig . 6 ) , which indicates that the T433 is searching for a best position to attack the Mg2+ in the complex . At the binding transition stage , T433 begins to attack the Mg2+ . The invasion pushes one of the coordinated waters , WAT3 , close to its neighbor WAT2 , which forces WAT2 to leave the stable coordination position with the Mg2+ cation ( Fig . 6 ( C ) ) . As we can see from Fig . 6 ( A ) , there is a steep decrease in the distance between T433 and Mg2+ together with a sharp increase of the distance between WAT2 and Mg2+ . On the other hand , the distance variation between WAT3 and Mg2+ is subtle ( Fig . 6 ( B ) ) . The stable coordination distance between Mg2+ and ligand is ∼2 . 0 Å . The coordination transition indicates that hydration waters may not necessarily be stripped at once . As in the case of F1-ATPase , the ATP may progressively exchange its hydrogen bonds with the hydration waters for hydrogen bonds with the ATP-pocket residues [8] . When the ATP-Mg2+ complex diffuse near the pocket , the negatively charged phosphate group will interact with positively charged or polar amino acids , such as Arginine ( R540 , R498 ) , Lysine ( K418 , K419 ) and Asparagine ( D502 , D474 ) . However , at the beginning , these charged groups may form hydrogen bonds with waters . When the ATP-Mg complex comes in , the waters may act as temporary bridges that should be weakened and broken with molecular vibrations during ATP-Mg2+ binding , and eventually be replaced and expelled by the ATP-Mg2+ complex . However , some of these water molecules will act as bridges via hydrogen bonds between the charged amino acids and the ATP-Mg2+ during the entire binding process . The 2 . 0 Å crystal structures of LTag in different nucleotide bound states also reveal some of these fixed water molecules in the binding pocket before and after ATP-Mg2+ binding . Here , we focus on the apical water and the water molecules coordinated with Mg2+ since they are directly related to the hydrolysis of ATP . Our experimental result show that the apical water is unusually coordinated through four residues: two cis-residues D474 , N529 and two trans-residues tR540 and tR498 ( Fig . 6 ( D ) ) . There is no particular order of coordination observed during the binding procedure . The distance between the four coordinated residues and the oxygen of the γPi group varies until the shrinking stage , when all the coordination distances converge to a stable hydrogen bond distance around 3 . 5 Å . The coordination procedure could be considered as a shrinking cage for the apical water ( Fig . 6 ( D ) ) . The vibration of the water molecule decreases until the cage shrinks to the stable state , at which point the apical water will be in a position ready for the nucleophilic attack in ATP hydrolysis . The PDLD/S-LRA method is used to evaluate the binding energy of a series of 20 key snapshots ( intermediate structures ) sampled from the TMD simulation trajectory . The results give a rough binding pocket energy profile between the Mg-ATP complex and binding pocket . Fig . 7 ( A ) and ( B ) show results using dielectric constants of 20 and 40 respectively . The calculated trends do not depend on the choice of protein dielectric constant . The energy profile starts at −6 kcal/mol , which corresponds to the interaction between Mg-ATP complex and the water from the beginning . And we use −6 kcal/mol as a base line to measure the binding energy . There is an energy barrier of 8 kcal/mol from WS to TS ( Fig . 7 ( A ) ) . The time corresponds to the Mg2+ coordination exchange , where the WAT2 ( Fig . 6 ( C ) ) escapes from its stable position due to the invasion of residue T433 . The coordination transition is similar to the transition from the Mg-ATP diphosphate coordination state and Mg-ATP tri-Phosphate coordination state . In the diphosphate coordination state , the Mg2+ coordinates with ATP through β and γ phosphates . In the triphosphate coordination state , the Mg2+ coordinates with ATP through all three phosphates . The transition energy barrier is around 11 kcal/mol ( 18 KbT ) in the water [18] , slightly larger than our simulation results . One possible explanation is that the conformation of the binding pocket protein may facilitate the coordination transition of Mg2+ by decreasing the barrier about 3 kcal/mol . This is followed by an energy valley of 13 kcal/mol , which lasts throughout the binding transition stage , and ends at the beginning of the shrinking stage . Then comes another energy barrier of 5 kcal/mol . There are three hydrogen bonds formed with N529 at this time ( Fig . 4 ) . One bond is formed with the γPi group oxygen , the other is formed with the tR498 and the third is formed with the apical water . The adjustment in the shrinking stage helps to prepare the apical water to attack the γPi in the following ATP hydrolysis stage . The energy profile stablizes at −12 to −14 kcal/mol , and the binding energy is about 8 kcal/mol . The experimental result of the TNP adenine nucleotide analogues binding energy is −8 kcal/mol ( −33 kJ/mol ) [19] , which could be used as a reference of our simulation results . Therefore , we conclude that our simulation results fall within a reasonable range , in comparison with the previous studies [8] . The major energy barrier is in the binding transition stage . Most of the binding energy is released during the docking and binding transition stage . On the other hand , most of the domain scale conformational changes happen after the binding transition stage ( Fig . 5 ) . The sequence may imply that the domain scale is triggered by the ATP binding . Similar models have been reported , for example , in the F0F1-ATPase model , the energy transduction takes places during the binding transition stage as well [8] . Some recent studies of similar T7 helicase [20] , [21] also reported that the global conformational change is triggered either by ADP release or by ATP binding . The motor domain engages with DNA after ATP binding [16] . The above simulation results indicate that the domain wise conformational changes happen in the ATP binding transition and the shrinking stages . The most significant conformational change is the D2/D3 domain movement towards the D1 domain ( or D2/D3 folding ) . The major folding movement occurs in the shrinking stage of the ATP binding process , with ∼20% occuring in the ATP binding transition stage . Our previous work has illustrated a ∼17 Å movement on the tips of the β-hairpin [3] . In this simulation study , we found that these two movements can be derived from the angled D2/D3 folding movement toward the N-terminal D1 domain , with an angle of approximately 20° . And the hinge point for the angled folding movement is around the joint of helix h5 and h6 ( Fig . 8 ( A ) ) . From the bottom view ( Fig . 8 ( C ) ) , the folding pushes the ATP-interacting cis-residues in an anti-clockwise direction to the neighboring trans-residues to form the cross-lock interactions to lock the ATP in the binding pocket . Fig . 8 ( C ) illustrates an interesting movement of the β-hairpin during the folding . The tip of the β-hairpin moves upward along the central channel with a screw motion , which is consistent with the simulation results in section 1 . Because the LTag monomer conformational changes triggered by ATP binding occur in the context of a hexamer , the six monomers within a hexamer have to cooperate with each other during the conformational switch . Our simulation shows that the most significant cooperative movement is the formation of the ATP binding pocket and the concomitant domain-wise folding of D2/D3 in the first transition stage . The cis-residues for ATP binding sit in the front and face towards the folding direction . The folding movement pushes the cis-residues to the position with the shortest locking distance ( bonding distance ) with the corresponding trans-residues of its anti-clockwise neighbor monomer for ATP binding . At the same time , the folding movement of the neighboring monomer slides the trans-residues , which are located at the right side of the monomer ( Fig . 8 ( E ) , to the contacting position for the incoming cis-residues . At the end of the cooperative movement , the two sides of the ATP binding pocket reach the shortest bonding distance to form the cross-locking interactions for ATP binding ( Fig . 8 ( E ) ) . Accompanying the folding , six β-hairpins rotate and move along their slant axes as illustrated in Fig . 8 ( D ) and Fig . 8 ( F ) . The ATP binding pocket is located at the base of the β-hairpin , thus the folding movement triggered by ATP binding could be amplified through the lever effects of six β-hairpins and transferred to the tip residues . The binding of six ATPs is therefore coupled with both the screw movements of the six β-hairpins towards the N-terminal in the central channel and the collective angle folding movement of the six D2/D3 domains towards the D1 domains , like an iris of the camera ( Fig . 8 ( E ) ) . However , we could not perform reliable computational analysis of the nature and extent of the cooperativity between the subunits within a hexamer at this time due to the lack of the experimental kinetics data on the cooperativity of LTag helicase .
ATP binding and hydrolysis by the LTag helicase motor is essential . We have performed a simulation study of the ATP binding process by LTag helicase in order to understand the energetics of ATP binding and the associated conformational changes for LTag helicase function in DNA unwinding . Based on our simulation results , we propose a cross-locking model for the ATP binding procedure for LTag helicase . The binding model can be divided into three main stages , namely , the docking stage from Apo state to the weak binding state , the binding transition stage from weak binding state to tight binding state , and the shrinking stage from the tight binding state to the ATP bound state . The first two binding stages are similar to the binding zipper model of the F1-ATPase system . During the ATP binding process , the Mg-ATP complexes diffuse to the binding pocket in the docking stage . And the phosphate group begins to interact with the binding pocket residues , such as the P-loop , and forms the conformation of WS . In the WS conformation , the bonding interactions between the three pairs of lock residues are not formed , and the three locks are fully open and the adenine group is completely outside the pocket . WS progresses to the TS during the binding transition stage . The Mg-ATP complex progressively forms hydrogen bonds with the residues in the binding pocket through the phosphate group and the ribose . These interactions induce the conformational changes of both ATP and the lock residues around the pocket . For the ATP , as the adenine inserts into the hydrophobic gap between h9 and h13 , the dihedral angle between the adenine/ribose and the phosphate group increases about 150 degrees . The ATP also bends down to a right angle . For the binding pocket , the three locks close sequentially , first lock1 ( Ribose and LYS419 ) , then lock2 ( ASP429 and LYS418/419 ) , and finally lock3 ( ASP474 and ARG540 ) . The hydrogen bond analysis shows that the Mg-ATP complex first interacts with the P-loop and cis-residues , and then forms hydrogen bonds with the trans-residues . The major stablizing hydrogen bonds begin to form with the cis-residues on the P-Loop and N529 , then trans-residues tK418 and tR540 . This corresponds with the results of the mutation study [15] and the ATP binding observation from the F0F1-ATPase system [8] , [14] . The number of hydrogen bonds increases linearly in the binding process , which is consistent with the results of the zipper binding model [8] . The apical water is important for the nucleophilic attack in ATP hydrolysis . Our simulation shows the position of the apical water is stabilized during the shrinking stage . The intra-ring conformational change and the relocation of residues compress the “cage” space around the apical water , and after certain adjustments , the coordinated residues are stabilized near the apical water in the ATP binding stage . At the end of the binding transition , the gate of the ATP pocket is fully closed . Negatively charged side chains , such as ILE428 , ASP429 , and ribose bases , all gather outside the gate . This may help to prevent the approach and binding of the other ATP ( Fig . 3 ( TS ) ) . In the binding transition stage , all the significant movement is concentrated in the binding pocket . Only ∼20% of the domain-wise conformational changes occur at this stage , which includes a subtle D2/D3 domain movement . The major domain-wise conformational changes ( ∼80% ) is accomplished in the shrinking stage . It is interesting to note that the radius of the central channel in the C-terminal bottom portion decreases more than the middle portion during the ATP-binding triggered conformational change , which could mean that the D2/D3 upwards movement toward the N-terminal D1 domain may generate a pushing force for moving DNA through the central channel . This movement is part of the iris-motion of LTag hexamer associated with the ATP binding and hydrolysis processes .
All the simulation is calculated using the CHARMM program package [22] and the binding energy profile is calculated by the POLARIS module of the Molaris program package [12] . The CHARMM27 all-atom force field [23] and the TIP3 water model [24] is employed . The cutoff radius for the non-bonded interactions is 14 Å . The SHAKE algorithm is adopted to fix the hydrogen bond during the simulation [25] . We have built two models for Apo state LTag helicase . The first Apo structure is built for TMD simulation of the ATP binding procedure . Six ATPs extracted from the ATP bound state are placed 20 Å away from the original Apo state helicase structure[3] . The O software program is used to adjust the ATP spatial position [26] . The ATPs are relaxed for 10 ps at 300 K . We use the Dowser program to place the internal water for the Apo structure [27] . A water sphere of 70 Å is built to wrap around the Apo structure . 36 chloride ions and 28 sodium ions are used to neutralize the system . We quenched the system for 10000 steps and then equilibrated for about 500 ps from 0 to 300 k , this is followed by another 200 ps equilibration for 300 K . The second model is built to verify the concerted model of ATP binding . The system is built by replacing one of the Apo state monomer with the corresponding ATP bound state structure . The position of the new monomer is decided by aligning the D1 domain to that of the original Apo state monomer . The system is quenched for 10000 steps and equilibrated for 500 ps from 0 to 300 k . The ATP bound state is scanned by the Dowser program to place the missing inner water . A 70 Å TIP3 water sphere wraps the system . Again , we quench the system for 10000 steps and equilibrated it from 0 to 300 K . The TMD simulation used an additional energy term based on the RMSD of the initial structure and final ( target ) structure . The energy term has the form: , where k is the force constant ( 20 kcal·mol−1·Å−2 ) , RMSD ( t ) is the root means square distance of the current simulated structure from the target structure , and RMSD* ( t ) is the predefined target RMSD value at time t . Since the forward and backward trajectory pathways are supposed to be the same , real crystal structure data provides a good starting point . Therefore , our TMD simulation started from the equilibrated ATP bound coordinates and ends at the equilibrated Apo state for 1 . 5 ns . The step size is 2 fs seconds . This strategy is also employed in the previous E . coli MurD study [28] . It is important to note that the Apo state monomer might not correspond to the ATP bound state monomer with the same segment name . We aligned each pair of the monomers between the Apo and the ATP bound state and save the 15 ( ) pair-wise alignment scores . Then align the six sequential monomers in the Apo state with those in the ATP bound state . For example , the segment of Apo and ATP bound state are represented by ABCDEF and A′B′C′D′E′F′ respectively . We first align the monomer sequence ABCDEF with A′B′C′D′E′F′ , and then align it to B′C′D′E′F′A′ , and next to C′D′E′F′A′B′ , and so on so forth . The final alignment is the one with the best overall sequence alignment score . In our study , we used the last 1 ns from the trajectory . The extra 0 . 5 ns is removed since it is related with the surface diffusion when ATP approaching the Apo helicase , which is out of the current research . To consolidate our results we have tried another two TMD simulations . One is the normal pathway from the equilibrated Apo state to the equilibrated ATP bound state . The conformational change is similar to the results above . Another TMD simulation involves an intermediate Apo state with ATPs bound to the pockets . The ATPs' positions are decided by aligning the ATP monomer with the Apo monomer . The intermediate Apo state is quenched and equilibrated in the same way as described above . The TMD simulation starts from the ATP bound state , and goes through the intermediate Apo state and ends at the Apo state . The simulation results are similar to the results presented above and therefore strengthen the results of our conformational pathway . The PDLD/S-LRA ( Linear Response Approximation version of the semi-microscopic PDLD ) method is designed to effectively evaluate the protein-ligand binding free energies through a thermodynamic cycle that is a fast approximation of the rigorous Free Energy Perturbation ( FEP ) [29] . The PDLD methods have been described in a series of theoretical papers , including the PDLD method [30] , the semi-empirical version , PDLD/S [31] and the fast approximation version , PDLD/S-LRA [32] . PDLD methods have been widely applied in the related biological systems , such as the F1-ATPase [33] and HIV protease [29] . Recently , we have also successfully applied the PDLD methods on the LTag DNA translocation analysis [34] . We used the PDLD/S-LRA method to evaluate the 20 snapshots ( intermediate structures ) from the simulated TMD trajectory . The PDLD/S-LRA method evaluates the change in electrostatic free energies upon transfer of a given ligand ( l ) from water to the protein by starting with the effective PDLD potentials; ( 1 ) where ΔGsol denotes the electrostatic contribution to the solvation free energy of the indicated group in water ( e . g . , denotes the solvation of the protein-ligand complex in water ) . The values of the ΔGsol's are evaluated by the Langevin dipole solvent model . is the electrostatic interaction between the charges of the ligand and the protein dipoles in vacuum ( this is a standard PDLD notation ) . This approach provides a reasonable approximation for the corresponding electrostatic free energies: ( 2 ) where the effective potential is defined in Eq . 1 and and designate an MD average over the coordinates of the ligand-complex in their polar and non-polar forms . It is important to realize that the average of Eq . 2 is always done where both contributions to the relevant are evaluated at the same configurations . That is , the PDLD/S energies of the polar and non-polar states are evaluated at each averaging step by using the same structure . The 20 structures are sampled evenly from the initial docking stage , through the binding transition stage ( WS to TS ) , and the shrinking stage ( TS to ATP bound state ) . All 20 structures are relaxed for 500 ps at 300 K . Each structure is then evaluated for 10 different runs . The mean values of these 20 structural evaluations are connected as a rough energy profile ( Fig . 7 ) .
|
The Large Tumor antigen ( LTag ) encoded by Simian Virus 40 ( SV40 ) is a marvelous molecule that is not only a viral oncogene , but also an efficient molecular machine as a helicase that unwinds double helix DNA for genome replication , an essential process in all living organisms . LTag hexameric helicase uses the energy of ATP to power its conformational switch for DNA unwinding . Understanding how the LTag conformational switch is coupled to the energy from ATP usage by LTag to do the mechanical work of unwinding DNA is of great interest to biologists , and yet remains to be established . Based on our previous high-resolution structures of LTag helicase in different conformational states , we simulated an LTag conformational transition pathway in the ATP binding process using the targeted molecular dynamics method . Our simulation results suggest a three-step process for the ATP binding to the nucleotide pocket , in which ATP is eventually “locked” into the pocket by three pairs of “locker” interactions . We have also quantitatively evaluated the energy profile of ATP binding using a special computational simulation technique . Additionally , our simulation study of ATP binding by LTag and the accompanying conformational switches in the context of a hexamer leads to a refined cooperative iris model that may be used for DNA unwinding .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"biology/protein",
"structure",
"prediction",
"computational",
"biology/molecular",
"dynamics"
] |
2009
|
A Computational Analysis of ATP Binding of SV40 Large Tumor Antigen Helicase Motor
|
While the circumstances surrounding the origin and spread of HIV are becoming clearer , the particulars of the origin of simian immunodeficiency virus ( SIV ) are still unknown . Specifically , the age of SIV , whether it is an ancient or recent infection , has not been resolved . Although many instances of cross-species transmission of SIV have been documented , the similarity between the African green monkey ( AGM ) and SIVagm phylogenies has long been held as suggestive of ancient codivergence between SIVs and their primate hosts . Here , we present well-resolved phylogenies based on full-length AGM mitochondrial genomes and seven previously published SIVagm genomes; these allowed us to perform the first rigorous phylogenetic test to our knowledge of the hypothesis that SIVagm codiverged with the AGMs . Using the Shimodaira–Hasegawa test , we show that the AGM mitochondrial genomes and SIVagm did not evolve along the same topology . Furthermore , we demonstrate that the SIVagm topology can be explained by a pattern of west-to-east transmission of the virus across existing AGM geographic ranges . Using a relaxed molecular clock , we also provide a date for the most recent common ancestor of the AGMs at approximately 3 million years ago . This study substantially weakens the theory of ancient SIV infection followed by codivergence with its primate hosts .
More than 30 nonhuman primate species in sub-Saharan Africa are naturally infected with simian immunodeficiency virus ( SIV ) [1]; however , the evolutionary forces shaping SIV diversity remain unclear . One of the most important unanswered questions regarding SIV evolution is whether it is an ancient infection that has been codiverging with its primate hosts for millions of years , or whether the virus may have arrived more recently and swept across already established primate lineages . Codivergence of viruses with their hosts has been inferred in other cases [2 , 3] , including other retroviruses [4 , 5] , where a close match between the host and viral phylogenetic trees suggests an ancient association . Furthermore , recent genomic analysis suggests that endogenous lentiviruses may have been infecting mammals for the last 7 million years [6] . Although it now seems clear that the overall pattern of the SIV and host phylogenies cannot be reconciled with a simple history of codivergence [7] , certain groups of SIVs and their hosts seem to suggest a shared evolutionary history . Among the SIV taxa , perhaps the best candidate for codivergence is the African green monkey ( AGM ) clade and their viruses , SIVagm . The AGM genus , Chlorocebus , consists of four species ( C . aethiops , C . pygerythrus , C . sabaeus , and C . tantalus ) , each with its own corresponding SIV lineage ( SIVgri , SIVver , SIVsab , and SIVtan ) [8–11] . The monkeys are geographically distributed across sub-Saharan Africa , with C . sabaeus in West Africa , C . tantalus in central Africa , C . aethiops ( grivet ) in northeastern Africa , and C . pygerythrus ( vervet ) ranging from East to southern Africa [12] . Studies using mitochondrial 12s rRNA have demonstrated monophyly among most AGM species ( i . e . , each individual shares a common ancestor more recently with every member of its own species than with any other AGM species ) [13 , 14] . However , 12s analysis provides very low statistical support for the branching order among the AGM taxa , and these studies were unable to resolve whether C . pygerythrus from Tanzania and South Africa are monophyletic or paraphyletic . On the face of it , the fact that this monophyletic clade of primates is infected by SIVs that also form a monophyletic clade provides compelling evidence of codivergence; however , a degree of caution is warranted whenever such inferences are made . An alternative mechanism by which pathogen and host topologies could resemble each other is preferential host-switching [7] . This model proposes that viruses are more likely to be transmitted between hosts with less phylogenetic distance separating them . This will lead to a viral phylogeny that is similar to the host tree , even in the absence of shared history . There is ample evidence demonstrating that SIV can switch hosts , with many examples of natural cross-species transmission of SIV among primates . SIVagm has been transmitted to the closely related patas monkey [15] and the more distantly related yellow and chacma baboons [16 , 17] . Furthermore , two distinct viral lineages infecting chimpanzees ( and possibly gorillas ) [18 , 19] and sooty mangabeys [20] have been introduced into the human population at least 11 times , giving rise to HIV [21] . In captivity , SIVagm has been transmitted to the African white-crowned mangabey [22] , and SIV from sooty mangabeys has been transmitted to several macaque species [23 , 24] . The relationships among SIVs are further complicated because many viruses , such as those infecting chimpanzees , sabaeus monkeys , mandrills , and Dent's Mona monkeys , represent recombinant lineages whose origins must have involved cross-species transmissions of SIV [25–28] . Nevertheless , additional evidence in favor of AGM–SIVagm codivergence has been put forward . The codivergence hypothesis predicts not only that the AGM species will share closely related SIVs , but also that the branching order within the virus clade and monkey clade should match . Such congruence has been reported from an analysis of the AGM CD4 gene [29] , which suggested phylogenetic congruence between this nuclear marker and the SIVagm env gene . However , the trees inferred for both virus and host genes were not well supported . Another study involving a nuclear gene , CCR5 , which codes for a coreceptor SIV uses to gain entry into host cells , concluded that coevolution between SIV and AGMs had occurred , implying an ancient infection [30] . More generally , the fact that primates naturally infected with SIV do not normally develop immunodeficiency seems to indicate a lengthy host–virus association . Prevalence of the virus in adult AGMs has been documented in excess of 70% [31 , 32] . Despite continuous viral replication , which can reach titers comparable to those found in humans infected with HIV [33 , 34] , immunodeficiency has only been observed once in an AGM that was co-infected with another retrovirus , STLV-I [35] . On the other hand , SIVagm is lethal when transmitted to non-African host monkeys such as the pigtailed macaque [36 , 37] . The low virulence observed in the natural host ( AGMs ) , however , does not necessarily indicate millions of years of evolution in response to SIV infection . Fossil evidence and genetic diversity studies propose that the AGM clade is on the order of millions of years old [38 , 39] , whereas molecular clock calculations have inferred a date of the most recent common ancestor ( MRCA ) of SIVagm at only hundreds or thousands of years old [40] . Estimates of such a recent origin of SIVagm cannot be dismissed simply on the basis of the observation that SIVagm is relatively benign in its natural hosts . The purpose of this study was to perform a rigorous phylogenetic test of the hypothesis of ancient codivergence between the AGMs and their SIVs . To do so , we sequenced complete AGM mitochondrial genomes , an approach that has produced what is , to our knowledge , the first statistically well-resolved AGM phylogeny . In comparing this phylogeny to ones inferred from SIVagm genomes , we found that the viral genome topology and host mitochondrial topology were incongruent and therefore provided no support for an ancient infection followed by codivergence .
We constructed the maximum likelihood ( ML ) SIVagm phylogeny using four previously published SIVagm genomes , one from each named species . The inferred phylogeny placed SIVgri ( grivet ) and SIVver ( vervet ) together with high bootstrap support ( Figure 1A ) . Midpoint rooting indicated that SIVsab was the most basal taxon . To test the robustness of our ML topology , we performed the Shimodaira–Hasegawa test ( SH-test ) [41] on all three possible unrooted SIVagm topologies . Using this conservative test , we were able to reject both alternative SIVagm unrooted topologies ( p < 0 . 05 ) ( Table 1 ) . To ensure that this pattern of SIVgri and SIVver forming a monophyletic clade was consistent for a larger sample of SIVagm strains , we also constructed a phylogeny using all seven available SIVagm genomes plus the complete env gene from two SIVver taxa that were isolated from C . pygerythrus in South Africa . We decided to include these subgenomic sequences because the complete SIVver genomes were all isolated from C . pygerythrus from East Africa , and we desired a better geographic representation of SIVver samples . Using this dataset , we recovered the same species-specific topology , with SIVgri and SIVver clustering together with strong support and SIVtan falling basal when rooted with SIVsab ( Figure 1B ) . All SIVver taxa form a monophyletic clade . To determine if available sequence data were sufficient to infer the branching order among the AGM species , we constructed phylogenies using the CD4 and CCR5 genes . Although available 12s rRNA data have proven useful for differentiating AGM species , they were not sufficient for resolving the phylogeny with statistical confidence . Furthermore , additional nuclear gene data have accumulated recently but have not yet been subjected to phylogenetic analysis . Despite earlier studies with fewer sequences , which seemed to determine the AGM topology , our results with the most complete alignments of nuclear gene sequences indicated that coding nuclear loci do not sufficiently resolve the AGM phylogeny . According to the CD4 topology , AGM species are not reciprocally monophyletic ( Figure 2A ) . There is low bootstrap support across the entire CD4 tree . We were also unable to resolve the branching order using CCR5 ( Figure 2B ) . All AGM species for which more than one CCR5 allele was analyzed exhibited paraphyly . Moreover , the only CCR5 allele from C . tantalus is identical to one of the C . sabaeus alleles , implying that CCR5 is not useful in distinguishing AGM species , let alone their phylogenetic relationships . To generate a sequence alignment likely to have sufficient phylogenetic signal to resolve the AGM phylogeny with a high degree of confidence , we sequenced complete mitochondrial genomes—an approach that has yielded robust phylogenies for other primates [42]—for C . sabaeus , C . tantalus , and C . pygerythrus from Tanzania and South Africa . Using an ML framework , we constructed a phylogeny comprised of these four genomes plus the previously published C . aethiops and C . sabaeus mitochondrial genomes . We inferred a single best topology that placed C . aethiops and C . tantalus together with high bootstrap support ( Figure 3 ) ; however , we were unable to resolve the phylogenetic relationship between the two C . pygerythrus taxa . While the ML tree indicated these two taxa are paraphyletic , with the taxon from South Africa branching off before the one from Tanzania , there is low bootstrap support for this inference . Of interest , the corrected genetic distance ( GTR + Γ4 ) between the two C . pygerythrus taxa was greater than that between C . tantalus and C . aethiops . Both midpoint and outgroup rooting using additional mitochondrial genomes ( Figure 4 ) placed C . sabaeus as the most basal AGM taxon . A topology identical to the one inferred via ML in PAUP* was also inferred in a Bayesian framework . This tree placed C . aethiops and C . tantalus together with a posterior probability of 1 . 0 . We then compared the unrooted AGM topologies using the SH-test , which rejected all AGM mitochondrial topologies that do not place C . tantalus with C . aethiops and C . sabaeus with C . pygerythrus ( Table 1 ) . However , we were unable to reject either of the alternate arrangements within C . pygerythrus , in which the Tanzanian C . pygerythrus branched first before the taxon from South Africa , or where the two C . pygerythrus taxa formed a monophyletic clade . Using a penalized likelihood approach [43] , we estimated the age of both the AGM MRCA and the subsequent radiation of C . aethiops , C . tantalus , and C . pygerythrus ( Figure 4 ) . This analysis was performed using trees generated under ML and Bayesian Markov-chain Monte Carlo ( MCMC ) frameworks . According to the ML analysis , the AGM lineages shared an MRCA 2 . 81 ± 0 . 35 million years ago ( MYA ) ; C . aethiops , C . tantalus , and C . pygerythrus shared an MRCA 1 . 48 ± 0 . 16 MYA . Estimates from the MCMC analysis did not differ considerably , placing the AGM common ancestor at 2 . 76 ± 0 . 23 MYA and the radiation at 1 . 59 ± 0 . 14 MYA . Because uniform branching order among the C . aethiops , C . tantalus , and two C . pygerythrus lineages was not observed in the ML analysis , we were unable to estimate the date of divergence events among these species . Our dates of other divergence events among the catarrhine species did not differ appreciably from those presented by Raaum et al . [42] and are therefore not reported here . As an explicit test for host–viral phylogenetic congruence , which , if present , would be a strong indication of codivergence , we used a series of SH-tests to determine if the SIVagm and AGM mitochondrial phylogenies were significantly different from each other ( Table 1 ) . First , we compared the ML SIVagm topology to the SIVagm topology that corresponded to the ML AGM mitochondrial topology ( labeled footnote b in Table 1 ) . The SH-test on the SIVagm genomes rejected this alternate topology ( p < 0 . 05 ) . Hence , there is convincing evidence that the SIVagm genomes did not evolve along the same topology as the AGM mitochondrial genomes . Of note , the test also rejected the alternate SIVagm topologies when the first 3 , 500 bases , the recombinant region in SIVsab , were included ( p < 0 . 05 ) ; thus , these results are not affected by the recombinant origin of SIVsab . Finally , we compared the ML AGM mitochondrial topology to the three AGM mitochondrial topologies that corresponded to the ML SIVagm topology ( labeled footnote c in Table 1 ) . We made these three comparisons because of the ambiguity that exists in the branching order of the two C . pygerythrus taxa . All three of these topologies were rejected by the SH-test on the AGM mitochondrial dataset ( p < 0 . 05 ) . In other words , we can confidently reject the hypothesis that the AGM mitochondrial genomes evolved along the same topology as the SIVagm genomes .
Our results present a significant challenge to the ancient origin of SIVagm followed by codivergence with their AGM hosts . Using an ML framework , we inferred robust phylogenies from the AGM mitochondrial genomes and SIVagm genomes . Although C . sabaeus are the most basal taxa for both the mitochondrial genome and SIVagm trees ( according to midpoint rooting methods ) , the other taxa do not share the same topology . As in previous studies on AGM taxonomy , we were unable to determine if C . pygerythrus from Tanzania and South Africa form a monophyletic clade , even though they are infected by the same SIVagm . Given the genetic distance between them , if C . pygerythrus is monophyletic , then it exhibits greater genetic diversity than is observed between C . tantalus and C . aethiops . In any case , using the SH-test we can confidently state that the virus did not evolve along the topology of the host mitochondrial genomes , and conversely , that the mitochondrial genomes did not evolve along the viral topology . These results demonstrate the usefulness of complete mitochondrial genomes in resolving recent primate divergence events , even those that occurred within a short span of time as with C . pygerythrus , C . tantalus , and C . aethiops . We also present the first date for the diversification of AGMs that accounts for the rate variation across the catarrhine phylogeny . The dating of the AGM MRCA at 2 . 81 ± 0 . 35 MYA indicates that if SIVagm did codiverge with its hosts ( which seems unlikely given our findings ) , it must have infected the AGM common ancestor nearly 3 MYA . Without evidence for a shared history , a model other than codivergence is needed to explain the observed pattern of SIVagm infection . A preferential host-switching model [7] , whereby viral transmission occurred over already established primate host ranges and favored the cross-species transmission of SIV from an initial AGM population to others , is a strong candidate . When the SIVagm phylogeny is mapped onto the distribution of AGMs in Africa , a geographic pattern of west-to-east transmission emerges ( Figure 5 ) . SIVsab is likely the most basal of the SIVagm taxa [1] , and its host , C . sabaeus , has the westernmost geographic distribution . SIVtan branches off next , and the range of C . tantalus begins at the Volta River , just east of the current C . sabaeus range , and continues into central Africa [12] . Finally , SIVgri and SIVver are the most derived SIVagm taxa and infect monkeys inhabiting the easternmost part of the continent . Since SIVagm is predominantly a sexually transmitted virus [32] , and AGM species are known to hybridize in the wild [12] , sexual encounters between AGM species may have facilitated SIV transmission at the edges of AGM ranges and the subsequent geographic spread of the virus . While geography and behavior likely provided ample opportunity for the transmission of an initial SIVagm variant from one AGM species to the next , these factors alone cannot explain why the various AGM species would have acquired SIV only from other AGM species , rather than other infected primate species . We speculate that intrinsic immunity factors , such as the APOBEC proteins [44–46] and TRIM5α[47] , may have played a role in this context . These proteins have been shown to prevent initiation of retroviral infection via a variety of mechanisms [48] . Specifically , these factors may have prevented the more distantly related Cercopithecus monkeys from becoming infected with SIVagm and similarly blocked the introduction of SIV from non-AGM species into the AGMs . In this light , the infection of the patas monkey with SIVsab might be an important clue , since this monkey is very closely related to the AGMs [49] and has been observed engaging in aggressive behavior with C . sabaeus [15] . Although the findings presented here on the age of SIV suggest that the virus was not a relevant force in the ancient evolution of these proteins , intrinsic immunity factors may have been crucial in shaping the distribution of SIV across the range of the African primates it infects . It is important to bear in mind that mitochondrial genomes , despite their length and phylogenetic information , represent only a single maternally inherited genetic locus . Regions of the AGM nuclear genome may have evolved along different evolutionary trajectories , some of which might be congruent with the viral evolutionary history . Furthermore , the AGM mitochondria may have experienced incomplete lineage sorting during the speciation events , which could obscure the species tree [50 , 51] . Nevertheless , our results represent the first and only statistically robust species-level AGM phylogeny to our knowledge , and this phylogeny unequivocally disagrees with the viral phylogeny . While there are examples of incongruence among mitochondrial and nuclear markers in the African guenons [52] , it is unlikely that other population level phenomenon such as introgression would occur in the mitochondria of the AGMs . The strong philopatry observed in AGM females , coupled with a dominance hierarchy that discourages breeding with migrant females , would decrease the likelihood of reproductive success of a female immigrant and therefore the probability of mitochondrial introgression [49] . In future studies , the inclusion of additional AGM mitochondrial genomes and other informative nuclear loci would be useful in determining if any of these population level phenomena have obscured the evolutionary history of the AGM . Nevertheless , we are confident that this study poses a significant challenge to the theory of ancient infection and codivergence . Given the conflicting AGM mitochondrial and SIVagm topologies presented here , the case for codivergence between AGMs and their SIVs is limited to the observations that ( 1 ) C . sabaeus and SIVsab are the basal taxa in the mitochondrial and SIV phylogenies , respectively , and ( 2 ) SIVagm forms a monophyletic group . However , the fact that C . sabaeus is basal in the mitochondrial phylogeny can hardly be used to argue in favor of codivergence when the remainder of the host phylogeny differs significantly from the viral one . In light of our findings , the ancient codivergence model is , to us , a less parsimonious explanation of the observed patterns than a preferential host-switching model with a relatively recent origin of SIVagm . In the absence of evidence in favor of AGM–SIVagm codivergence , we are left to wonder about the case for codivergence in other African monkeys infected with SIV . A recent ancestry of SIVagm calls into question the conclusions put forth by Kuhmann et al . [30] regarding the coevolution of SIVagm with host protein CCR5 . Our analysis of the CCR5 locus suggests that there are no unique species-specific differences among the alleles that would suggest coevolution . Furthermore , the study by Kuhmann et al . did not perform a formal test for selection and assumed that a higher proportion of nonsynonymous-to-synonymous substitutions was evidence of positive selection; however , the ratio they observed ( approximately two nonsynonymous changes for every one synonymous change ) is consistent with purifying selection , as nonsynonymous mutations are more frequent by chance alone . If SIVagm is not the result of an ancient infection , then its avirulence in its natural hosts may have evolved over a much shorter time frame than implied by the ancient codivergence model . Competition experiments by Ariën et al . [53] between matched pairs of HIV samples from 2002–2003 and the late 1980s suggest that the virus may be attenuating in the human population . The authors proposed that this loss of replicative fitness by HIV might be due to its adaptation to the human immune system coupled with repeated bottlenecks resulting from human-to-human transmission . Their data suggest that evidence of reduced virulence could be perceived in relatively short periods of time . Precise dating of the original SIV infection in the AGMs may help us better appreciate the evolutionary time frame in which such change is possible in the viral lineage .
DNA extracts from C . pygerythrus from Tanzania ( CAE9649 ) , C . pygerythrus from South Africa ( V389 ) , C . sabaeus ( Letta ) , and C . tantalus ( Bébé ) were provided by A . C . van der Kuyl . Mitochondrial genomes were amplified using three PCR primer sets whose products ranged from 5 to 8 kilobases , which were designed based on the method developed by Raaum et al . [42] . Reactions were performed using the TripleMaster PCR system with an annealing temperature of 52 °C with an extension time of 9 min for the first ten cycles , which was extended 15 s for each additional cycle . Each reaction was run for 35 cycles . PCR products were purified using QIAquick PCR purification kits ( Qiagen , http://www . qiagen . com/ ) , and the templates were sequenced using internal primers . Regions that proved difficult to sequence from original template were re-amplified using internal PCR primers and then sequenced using those primers . All primer sequences are shown in Table S1 . PCR reactions were confirmed using a 0 . 8% agarose gel stained with SYBR Safe ( Invitrogen , http://www . invitrogen . com/ ) . DNA sequencing was performed by the Genomic Analysis and Technology Core Facility ( University of Arizona , Tucson , Arizona , United States ) using an automated sequencer ( Applied Biosystems 3730XL DNA Analyzer , http://www . appliedbiosystems . com/ ) until each base had been sequenced at least twice . Contigs were then assembled using Sequencher version 4 . 2 ( Gene Codes Corporation , http://www . genecodes . com/ ) . Each of the four mitochondrial genomes was completely sequenced , except for C . sabaeus , for which a 200–base-pair region proved problematic , possibly due to secondary structure of the template . In addition , the mitochondrial genome of C . tantalus exhibited a repeat structure within its control region that , while unusual , is not unprecedented [54] . A 115–base-pair region was repeated as many as three instances in some sequencing reactions , whereas in others it appeared only a single time . PCR amplifications of the C . tantalus control region indicated that multiple forms of this region existed . Unfortunately , due to degradation of our original template , we were unable to determine whether this repeat structure was due to PCR error or actual heterogeneity in the sample . Nevertheless , this region was excluded from our analysis , because the repeats had no homologous region in any other mitochondrial genome and were therefore phylogenetically uninformative . CD4 and CCR5 sequences were downloaded from GenBank . CD4 genes labeled as Barbados were classified as C . sabaeus based on recent genetic testing using cytochrome b sequence analysis [55] . All redundant CCR5 sequences were removed , except for those that were isolated from different species . Each of these datasets was aligned by hand using Se-Al [56] . ML phylogenetic trees for these two loci were inferred using a heuristic search in PAUP* version 4 . 0b10 [57] . The models of nucleotide substitution , Kimura81 + Inv for CD4 and HKY + Inv for CCR5 , were identified by ModelTest version 3 . 7 [58] . Bootstrap support was assessed using 1 , 000 and 100 replicates for the CD4 and CCR5 topologies , respectively , using the ML nucleotide substitution parameters estimated from the ML phylogeny . The four AGM mitochondrial genomes sequenced here and the previously published C . aethiops and C . sabaeus mitochondrial genomes were aligned by hand using Se-Al , except for the variable D-loop regions , which were aligned using CLUSTAL X [59] . A single phylogenetic tree was inferred using an exhaustive search with ML parameters inferred under a GTR + Γ4 nucleotide substitution model in PAUP* . Bootstrap support was assessed in an ML framework whereby the nucleotide substitution parameters were reestimated for each replicate and a heuristic search was performed; this was done for 1 , 000 replicates . In addition , a phylogenetic tree was inferred with a GTR + Γ4 nucleotide substitution model in a Bayesian framework using MrBayes version 3 . 0 [60] . Two independent runs were performed , each using 1 million steps with four chains sampling every 100 steps . The first 10% of the trees were removed and posterior probabilities were calculated from these post-burnin trees . SIVagm genomes were obtained from the HIV Sequence Database at Los Alamos National Laboratory ( LANL , http://hiv . lanl . gov/content/hiv-db/mainpage . html ) . In the initial analysis , the four genomes were aligned using CLUSTAL X . We excluded the first 3 , 500 bases of all SIVagm genomes from our analyses , because SIVsab is a known recombinant in the 3′ part of this region , and its phylogenetic placement is ambiguous in the 5′ section of this region [25] . The sequences were aligned using CLUSTAL X , and an exhaustive search inferred a single phylogenetic tree using ML parameters estimated under a GTR + Γ4 model in PAUP* . In the secondary analysis on all seven published SIVagm genomes and the env genes of SIVver from South Africa , the sequences were also obtained from the LANL database . A single phylogenetic tree was found using a heuristic search with ML parameters inferred under a GTR + Γ4 model in PAUP* . Bootstrap support was assessed in an ML framework whereby each nucleotide substitution parameter was reestimated for each replicate and a heuristic search was performed; this was done for 1 , 000 replicates for the four-taxa tree and for 100 replicates for the nine-taxa tree . The SH-test was performed in PAUP* on the unrooted bifurcating topologies for the six AGM mitochondrial genomes , in which the C . sabaeus taxa are monophyletic , and the four initial SIVagm genomes . The test parameters were estimated using a GTR + Γ4 model with 1 , 000 RELL replicates . Molecular clock analysis was carried out using the r8s software developed by Sanderson [61] . In order to estimate the divergence dates of and within the AGMs , we included other complete mitochondrial genomes from the Old World monkeys Colobus guereza , Macaca sylvanus , Papio hamadryas , and Trachypithecus obscurus; lesser and great apes Hylobates lar , Gorilla gorilla , Homo sapiens , Pan paniscus , Pan troglodytes , Pongo pygmaeus pygmaeus , and Pongo pygmaeus abelii; and a New World monkey , Cebus albifrons , which was used as an outgroup to root the phylogeny . An alignment of these mitochondrial genomes was obtained using CLUSTAL X . The two variable D-loop regions were removed from further analysis due to their poor sequence conservation . Our analysis closely followed that of Raaum et al . [42] , who first estimated divergence dates using many of the same primate mitochondrial genomes . We used a semiparametric approach with a penalized likelihood method in which the rate of evolution along each branch is allowed to vary , but a roughness penalty prevents the rate from varying too much from branch to branch [61] . An optimal smoothing parameter was chosen by cross-validation analysis . The non-clocklike behavior of this dataset was not unexpected given the decrease in the rate of evolution observed in apes [62 , 63] . We based our divergence estimates on three fossil-derived calibration points identified by Raaum et al . : the 6-MYA split between Pan and Homo , the 14-MYA split between the Asian great apes ( Pongo ) and the African great apes , and the 23-MYA split between hominoids and the Old World monkeys . These fossil-derived dates were entered into r8s as point estimates , rather than intervals , because r8s does not work well with narrow calibration windows . To estimate confidence intervals for the age of the AGM clade and the radiation of C . aethiops , C . tantalus , and C . pygerythrus , we used ML branch lengths estimated from 100 nonparametric bootstrap replicate trees in PAUP* and 100 trees from a Bayesian MCMC run . Bootstrap trees in PAUP* were obtained using GTR + Γ4 parameters estimated from an ML tree . Trees from the MCMC run were sampled every 9 , 000 trees after the first 100 , 000 burnin trees . In both cases , every tree supported the identical topology for all taxa except C . aethiops , C . tantalus , and the two C . pygerythrus . The Bayesian analysis did , however , place C . aethiops and C . tantalus together 100% of the time , which is consistent with our previous phylogenetic analysis on the AGM mitochondrial genomes . We provide estimates of error as two standard deviations from the mean age of the estimated node for each of these datasets ( ML and MCMC ) ; these estimates are conservative , as they do not capture the uncertainty in the fossil record .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/ ) accession numbers for the AGM mitochondrial genomes sequenced in this study are EF597500–EF59750 . Accession numbers for other genes and genomes are as follows: previously published AGM mitochondrial genomes ( AY863426 and DQ069713 ) , CCR5 ( AB015944 , AF035221 , AF035222 , AF035223 , AF081577 , AF105286 , AF162006 , AF162007 , AF162016 , AF162017 , AF162020 , AF162022 , AF162023 , AF162025 , AF162026 , AF162030 , AF162031 , AF252552 , U83324 , and U83325 ) , CD4 ( AF001221–AF001228 , D86589 , and X73322 ) , initial SIVagm genomes ( M66437 , L40990 , U58991 , and U04005 ) , additional SIVagm genes and genomes ( BD092095 , M30931 , M29975 , AF015905 , and AF015906 ) , and additional primate mitochondrial genomes ( AY863427 , NC_002764 , Y18001 , AY863425 , X99256 , NC_001645 , NC_001807 , NC_001644 , NC_001643 , NC_001646 , NC_002083 , and AJ309866 ) .
|
Elucidating the factors that influence the emergence of viral pathogens is of great importance to the study of infectious disease . HIV is understood to have originated from simian immunodeficiency viruses ( SIVs ) infecting nonhuman African primates , but the length of time the virus has been present in these apes and monkeys is not known . These infected primates do not normally develop immunodeficiency , and understanding the age of SIV might help explain why . It has been suggested that some of these monkeys have been infected for millions of years , because many closely related monkey species are infected with closely related viruses . One of the most prominent examples of this relationship is between the African green monkeys and their SIVs . In this study , we compared viral phylogenetic trees to those of their hosts' mitochondrial genomes and found that they do not support the theory of ancient infection followed by codivergence . Our results suggest that SIV did not infect these monkeys until after speciation and subsequently swept across their geographical ranges . If this infection is relatively recent , then avirulence may have evolved over a shorter time frame than previously suggested . This finding could have implications for the future trajectory of HIV disease severity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"evolutionary",
"biology",
"viruses",
"primates",
"virology"
] |
2007
|
A Challenge to the Ancient Origin of SIVagm Based on African Green Monkey Mitochondrial Genomes
|
Tetherin/BST-2/CD317 is a recently identified antiviral protein that blocks the release of nascent retrovirus , and other virus , particles from infected cells . An HIV-1 accessory protein , Vpu , acts as an antagonist of tetherin . Here , we show that positive selection is evident in primate tetherin sequences and that HIV-1 Vpu appears to have specifically adapted to antagonize variants of tetherin found in humans and chimpanzees . Tetherin variants found in rhesus macaques ( rh ) , African green monkeys ( agm ) and mice were able to inhibit HIV-1 particle release , but were resistant to antagonism by HIV-1 Vpu . Notably , reciprocal exchange of transmembrane domains between human and monkey tetherins conferred sensitivity and resistance to Vpu , identifying this protein domain as a critical determinant of Vpu function . Indeed , differences between hu-tetherin and rh-tetherin at several positions in the transmembrane domain affected sensitivity to antagonism by Vpu . Two alterations in the hu-tetherin transmembrane domain , that correspond to differences found in rh- and agm-tetherin proteins , were sufficient to render hu-tetherin completely resistant to HIV-1 Vpu . Interestingly , transmembrane and cytoplasmic domain sequences in primate tetherins exhibit variation at numerous codons that is likely the result of positive selection , and some of these changes coincide with determinants of HIV-1 Vpu sensitivity . Overall , these data indicate that tetherin could impose a barrier to viral zoonosis as a consequence of positive selection that has been driven by ancient viral antagonists , and that the HIV-1 Vpu protein has specialized to target the transmembrane domains found in human/chimpanzee tetherin proteins .
Eukaryotic cells can constitutively or inducibly express a variety of molecules that inhibit the replication of viruses . Among these antiviral defenses are components of the type-I interferon ( IFN ) -induced innate immune system [1] , [2] . In turn , viruses have evolved to express proteins that either limit IFN-induced gene expression or directly antagonize the function of antiviral proteins . We and others recently identified an IFN-induced antiviral protein , termed tetherin , that functions by a novel mechanism . Specifically , tetherin blocks the release of nascent virions from HIV-1 infected cells [3]–[5] . Tetherin is an integral membrane protein with a unique topology . In particular , it encodes a transmembrane anchor towards its N-terminus , as well as a putative glycophosphatidyl-inositol lipid anchor at its C-terminus [6] . These two membrane anchors are linked by an extracellular domain that is predicted to form a coiled-coil . Ectopic expression of tetherin in cells that do not ordinarily express it results in the formation of protease-sensitive tethers that causes retention of retrovirus particles on the surface of infected cells , from where they can be internalized [4] , [5] , [7] , [8] . This pronounced ability to retain and internalize HIV-1 particles is present constitutively in cells that normally express tetherin , but is suppressed when tetherin is depleted . Tetherin colocalizes with Gag and appears to act by inducing adherence of virion and cell membranes . Thus , virions that are retained by tetherin are fully formed and mature , and have lipid bilayers that are discontinuous with cell membranes [4] , [7] . Notably , an HIV-1 accessory transmembrane protein , Vpu , acts as a viral antagonist of tetherin [4] , [5] . Indeed tetherin dramatically inhibits the release of Vpu-defective HIV-1 virions , but has only modest effects on wild-type Vpu-expressing HIV-1 . Moreover , Vpu colocalizes with tetherin and prevents the localization of tetherin to nascent virions , perhaps through its ability to reduce the amount of tetherin at the cell surface [4] , [5] . Thus , the existence of tetherin explains the previously observed requirement for Vpu during HIV-1 particle release from certain cells , particularly those that have been exposed to type-I IFN [3] , [7] , [9]–[12] . The wide expression of tetherin upon exposure of cells to IFN-alpha [4] , [13] and the wide range of retroviruses and filoviruses that are inhibited by tetherin [8] suggests that it might be a general component of an innate immune defense against many enveloped viruses . As such , tetherin could provide an impetus for the evolution of antagonists in viruses other than HIV-1 . Indeed , the Kaposi's sarcoma herpesvirus ( KSHV ) also encodes a likely antagonist of tetherin , since expression of the KSHV K5 protein decreases the steady state level of tetherin protein [14] . Additionally , certain retroviral envelope proteins , in particular the HIV-2 Env , have Vpu-like activity [15] , [16] . Thus , it seems likely that tetherin antagonists other than Vpu exist . Here , we show that tetherin proteins from different species exhibit marked differences in sensitivity to antagonism by HIV-1 Vpu . Specifically , tetherin proteins from two old world monkey species as well as from mouse , are effective inhibitors of HIV-1 particle release , but are resistant to Vpu . Moreover , we show that the transmembrane domain of tetherin contains determinants of sensitivity to Vpu , and that two mutations in the human tetherin sequence are sufficient to generate a protein that is entirely resistant to antagonism by Vpu . Interestingly , tetherin sequences that are predicted to be accessible to cytoplasmic or integral membrane antagonists , such as HIV-1 Vpu , exhibit evidence of positive selection at numerous codons , including some that we demonstrate determine the effectiveness of Vpu antagonism . Thus , past selective pressures imposed on tetherin by viral antagonists likely provides a barrier to the establishment of zoonotic infections by modern primate lentiviruses , and potentially enveloped viruses from other species , whose spread depends on antagonism of tetherin function .
Inspection of sequence databases revealed the presence of putative tetherin proteins in various mammalian species . For functional analyses , we amplified tetherin sequences from cDNAs derived from rhesus macaque ( rh ) , African green monkey ( agm ) , chimpanzee ( cpz ) and mouse ( mo ) . Coexpression of hu-tetherin , cpz-tetherin , rh-tetherin , agm-tetherin , mo-tetherin , with HIV-1 ( delVpu ) caused a marked decease in the yield of viral particles , as measured using infectivity or western blot assays ( Fig . 1 A , B , C ) . The magnitude of the reduction in virus yield varied somewhat , depending which tetherin protein was coexpressed . In each case , inhibition of virion release by tetherin did not lead to a dramatic accumulation of cell associated viral proteins . This finding suggest that the virions that are retained by tetherin , are destroyed at rate that exceeds , or is not greatly different , to their synthesis . This would likely be through endocytosis , followed by lysosomal degradation . Moreover , it may simply be the case that only a fraction of the viral protein that is synthesized by infected cells is actually released as particles . Thus the amount of viral protein that is observed in cell lysates would be determined by its intrinsic turnover rate , rather than particle release versus retention . The differences in potency that were observed among the various tetherin proteins were only partly explained by variation in the levels at which each tetherin was expressed ( Fig . 1 , Fig . S1A ) , A finding which suggests that natural variation in potency may exist among mammalian tetherin proteins , and that hu-tetherin is particularly potent . However , a caveat to be attached to this conclusion is that transiently expressed tetherin exisits as a variety of species , presumably reflecting heterogeneous glycosylation . At present it is not clear whether all of the various tetherin species are active , and it is possible that the amount of active tetherin is not uniformly represented in analyses of total tetherin protein levels . Nonetheless , each tetherin protein was clearly capable of reducing infectious HIV-1 ( delVpu ) yield , by 20-fold or more . Strikingly , however , and in contrast to hu-tetherin , the rh- , agm- and mo-tetherin proteins were equally efficient in reducing HIV-1 ( WT ) and HIV-1 ( delVpu ) virion yield ( Fig . 1A , B , C ) . Conversely , as with hu-tetherin , HIV-1 ( WT ) was substantially resistant to inhibition by cpz-tetherin , while the release of HIV-1 ( delVpu ) particles was strongly inhbited ( Fig . 1 D , E ) . Thus , the non-hominid tetherin proteins were apparently insensitive to antagonism by HIV-1 Vpu . while both hu-tetherin and cpz-tetherin were Vpu-sensitive . Earlier work indicated that expression of an intact transmembrane ( TM ) segment of HIV-1 Vpu was most important , and perhaps sufficient , to enhance HIV-1 particle release [17] . This finding , combined with the notion that only the N-terminal portion of tetherin should be physically accessible to Vpu , suggested the possibility that Vpu might target the transmembrane domain of tetherin . Thus , we generated chimeric proteins , termed hu ( agmTM ) and hu ( rhTM ) , in which the transmembrane domain of hu-tetherin was replaced with corresponding sequences from rh- or agm-tetherin . These chimeric proteins differed in the magnitude with which they inhibited HIV-1 virion release , concordant with differences in expression level ( Fig . S1B ) , but both were entirely resistant to antagonism by Vpu ( Fig . 2A , C ) . Indeed , the hu-tetherin protein containing the agm TM domain inhibited both HIV-1 ( delVpu ) and HIV-1 ( WT ) particle release with similar or greater potency with which the intact hu-tetherin protein selectively blocked HIV-1 ( delVpu ) release ( Fig 2A , C ) . In a reciprocal experiment , rh-tetherin and agm-tetherin proteins encoding a hu-tetherin transmembrane domain were generated . The rh ( huTM ) protein was expressed at a slightly higher level and was a more effective inhibitor of HIV-1 particle release than was the agm ( huTM ) protein ( Fig . 2B , C , Fig . S1B ) . However , both proteins selectively inhibited HIV-1 ( delVpu ) particle release , indicating that they were sensitive to antagonism by HIV-1 Vpu . Thus , the exchange of the TM domain between tetherin proteins from different species transferred sensitivity and resistance to antagonism by HIV-1 Vpu . Inspection of tetherin TM domain protein sequences revealed several differences between the monkey and human proteins , distributed along the length of the TM domain ( Fig . 3A ) . To determine which of these were responsible for the Vpu resistance of the monkey tetherin proteins to antagonism by Vpu , we generated mutant forms of hu-tetherin , each bearing an individual change in the TM domain that is found at the corresponding position in rh-tetherin . ( The exceptions to this scheme were the L23V , L24I mutant in which two contiguous amino acids were changed to their rh-tetherin counterparts and the delGI change , where a two amino acid deletion that is present in the rh-tetherin sequence was introduced ) . This panel of mutant tetherin proteins varied in the potency with which they reduced HIV-1 ( delVpu ) virion yield , and in their ability to be antagonized by Vpu ( Fig . 3B , C ) . Variation in the potency of antiviral activity among the mutant panel correlated with expression level in most cases ( Fig . S1C ) . Notably , none of the individual hu-tetherin mutants recapitulated the phenotype of the rh-tetherin or hu ( rhTM ) -tetherin proteins that appeared completely resistant to Vpu antagonism . Rather , several of the individual mutants appeared partly resistant to antagonism by Vpu , in that Vpu was not able enhance virion release as effectively in their presence as it did in the presence of unmanipulated hu-tetherin ( Fig . 3B , C ) . Because these analyses were slightly confounded by the variation in the potency and expression of the individual tetherin mutants , we measured infectious HIV-1 ( WT ) and HIV-1 ( delVpu ) virion yield in the presence of varying levels of the mutant tetherin proteins ( Fig . 4 ) . Overall , these analyses identified a single amino acid difference ( P40L ) as contributing substantially to the Vpu sensitivity of hu-tetherin and the Vpu-resistance of rh-tetherin . Other changes in the hu-tetherin TM had more modest effects , or no effect on Vpu sensitivity , when present in isolation , or substantially affected tetherin potency ( Fig . 3 , Fig . 4 ) . Because our initial analysis revealed that no single change in the TM domain of hu-tetherin could abolish sensitivity to Vpu , we next tested whether mutations that individually had minor or partial effects on Vpu antagonism could exert more dramatic effects when present in combination . In particular , several mutations were combined with the most obvious difference between human and monkey tetherin TM domains , namely the delGI change . Notably , the delGI , T45I double mutant strongly inhibited HIV-1 particle release and was completely resistant to antagonism by Vpu ( Fig . 5A , B ) . Additionally , the I33V , I36L combination mutation which would be predicted to target proximal residues on the face of a TM alpha helix ( Fig . 3A ) appeared to confer at least partial resistance to antagonism by Vpu . However , this double mutation generated a tetherin protein with only modest activity . Nonetheless , when combined with a mutation at a third proximal residue ( generating V30G , I33V , I36L ) this combined mutation conferred partial resistance to Vpu antagonism , in the context of a protein with potent antiviral activity ( Fig . 5A , B ) . Moreover a hu-tetherin bearing combined delGI , I33V , I36V mutations was almost completely resistant to antagonism by Vpu , and exhibited substantial antiviral activity ( Fig . 5A , B ) . Finally , combining the delGI mutation with subsitiutions at contiguous residues that corresponded to rh-tetherin residues ( delGI , L23V , L24I ) resulted in a protein with only weak inhibitory activity , whereas combining the delGI mutation with contiguous L23A , L24V mutations ( corresponding to agm-tetherin residues ) generated a protein that was potent and partly Vpu resistant ( Fig . 5A , B ) . Notably , the differences in activity and Vpu sensitivity among the various combination-mutant tetherin proteins was not explained by differences in expression level ( Fig . S1D ) . To determine whether mutations that conferred resistance to antagonism by Vpu in virion release assays also conferred resistance to the previously described phenomenon of Vpu-induced downregulation of hu-tetherin from the cell surface [5] , we generated cell lines stably expressing either wild type hu-tetherin-HA protein or the Vpu-resistant delGI/T45I hu-tetherin mutant ( Fig . 6 ) . Upon infection with HIV-1 ( WT ) , hu-tetherin was efficiently depleted from the surface of the vast majority of infected cells ( Fig . 6A , B ) . Conversely , infection with HIV-1 ( delVpu ) resulted in little or no hu-tetherin downregulation from the surface of infected cells . Strikingly , and unlike the WT hu-tetherin protein , the delGI/T45I mutant hu-tetherin was not removed from the cell surface upon infection with HIV-1 ( WT ) ( Fig . 6A , B ) and , thus , was resistant to surface downregulation by Vpu . Overall , these experiments revealed that no single difference between the hu-tetherin and rh-tetherin proteins accounted for their respective sensitivity and resistance to antagonism by HIV-1 Vpu . Rather , they indicated that the particular combination of residues in the tetherin TM domain can affect antiviral potency , and that multiple differences between human and monkey proteins , including the delGI indel , and the I33V , I36L , P40L and T45I differences , influence the differential sensitivity of tetherins to antagonism by Vpu . Inspection of a larger collection of mammalian tetherin sequences amplified from various old world primates , or retrieved from sequence databases , revealed some striking features . Firstly , among the nonprimate mammalian tetherin sequences , the N-terminal cytoplasmic domain was hypervariable , both in length and sequence ( data not shown ) . Because of these properties , it proved impossible to unambiguously align non-primate and primate tetherin sequences in order to perform tests for positive selection . Therefore , we confined further analyses to primate tetherin sequences , which could be aligned unambiguously throughout the entire length of the coding sequence . Even within primates there was considerable sequence divergence between species , ranging from 0 . 5% to 40 . 0% at the nucleotide level . Sampling within three old world monkey species ( rhesus macaques , pig-tailed macaques and sooty mangabeys ) also revealed the presence of significant polymorphism within species ( Fig . S2 and data not shown ) ; it remains to be seen whether nonsynonymous polymorphism extends to the tetherin locus of other primate lineages . Positive selection was tested using the REL ( HyPhy ) [18] and CODEML ( PAML ) [19] methods and these analyses revealed that codons exhibiting high dN/dS ratios , and therefore likely to have been subjected to positive selection , were enriched in the N-terminal cytoplasmic and TM domains in primate tetherins ( Fig . 7 ) . Tetherin evolution in primates was also evaluated under several standard models of sequence evolution as implemented in the CODEML program . These comprise three nested pairs of models ( M0 and M3; M1a and M2a; M7 and M8 ) in which the second model of each pair is derived from the first by allowing sites to evolve under positive selection . Nested models were compared using the likelihood ratio test , and in each case allowing individual sites to evolve under positive selection ( M3 , M2a , M8 ) gave a significantly better fit to the primate sequence data than the corresponding model without positive selection ( M0 , M1a and M7 , respectively ) ( Table 1 ) . The M3 , M2a and M8 models identified a largely overlapping set of sites in the tetherin coding sequence with dN/dS>1 , consistent with an evolutionary history characterized by frequent episodes of positive selection . Notably , some codons that exhibited a high probability of having evolved under positive selection coincided with residues that determined the effectiveness of Vpu antagonism ( Fig . 7 ) . However , there were numerous additional codons , particularly in the tetherin cytoplasmic domain , that also exhibited high dN/dS ratios , suggesting that antagonists other than Vpu have also imposed selective pressure on primate tetherin sequences .
Previously , we reported that Vpu could reverse the inhibitory effect of IFN-alpha on HIV-1 particle release from human cells , but that Vpu failed to reverse such IFN-alpha induced inhibition in African green monkey cells [3] . The subsequent discovery that tetherin is an IFN-induced inhibitor that is antagonized by Vpu [4] , [5] leads to the prediction that agm-tetherin should be resistant to Vpu . Here we show that agm-tetherin , as well as rh-tetherin and mo-tetherin are indeed resistant to antagonism by Vpu , in contrast to tetherin variants found in species ( human and chimpanzee ) that are permissive hosts for HIV-1 . Moreover , these studies identify the TM domain of hu-tetherin as a major determinant of the effectiveness of Vpu antagonism . That the TM domain of tetherin harbors critical determinants of Vpu sensitivity is concordant with previous observations indicating that the TM domain of Vpu is critical for its virus release activity [17] . Thus , these observations suggest a model in which Vpu and tetherin interact via their TM domains . Such an interaction could be direct , but further work will be required to resolve whether this is indeed the case , and precisely how Vpu tetherin antagonism and/or downregulation from the cell surface is achieved is not known . In this regard , the recent report suggests that the host cell protein CAML is important for Vpu activity [20] , but how tetherin , Vpu and CAML interact in a functional sense is unclear at present . Importantly , no single change in the hu-tetherin TM domain abolished Vpu sensitivity , which is consistent with a model in which Vpu , or a bridging factor , makes multiple contacts with the TM domain of tetherin . Indeed , the hu-tetherin mutant , delGI , T45I , that had the most striking phenotype , in that it retained full activity but was completely resistant to antagonism by Vpu , harbored mutations at positions close to the opposing ends of the TM domain . Additionally , a different combination of mutations ( delGI , I33V , I36L ) also conferred complete Vpu resistance , again consistent with the notion that multiple contacts with the tetherin TM domain are made during its antagonism by Vpu . Concordant with these findings , the chimpanzee and human tetherin proteins were both sensitive to HIV-1 Vpu , and differed from each other at only a single position in their TM domains . Thus , it is likely only minor , if any , adaptation in the SIVCPZ Vpu protein that is immediately ancestral to the HIV-1 Vpu proteins would have been required in order for it to target human tetherin . Primate tetherin TM domain sequences , that should only be accessible to integral membrane antagonists , exhibit clear evidence of positive selection at several codons , including some that determine the effectiveness of Vpu antagonism . Thus , it is likely that antagonists encoded by pathogenic viruses have driven the selection of the tetherin variants that exist in modern primates . Such antagonists could include Vpu itself , since several primate lentiviruses encode Vpu proteins ( http://www . hiv . lanl . gov/ ) . However , other viral antagonists , including the KSHV K5 protein [14] , or homolgues of it , are also reasonable candidates for factors that have imposed selective pressure on tetherin sequences . In addition to the TM , several sites with the highest dN/dS ratios mapped to the N-terminal cytoplasmic domain . Such sites may define an accessible target exploited by other virally encoded inhibitors of tetherin . It is noteworthy in this regard that many primate lentiviruses do not encode a Vpu like protein , and may have evolved alternative strategies targeting this or other regions of the protein . Additionally , tetherin sequences might also have evolved to better target specific types of viral particles in some host species , as a consequence of varying viral challenges . Such evolution might also include adaptations in domains of tetherin that , for example , modify its trafficking within cells ( likely including the cytoplasmic and TM domains ) . Overall , there are several potential sources of evolutionary pressure that could give rise to positive selection and diversification of tetherin genes . We note that the sequence of the TM domain of tetherin is obviously constrained by the need to retain the biochemical characteristics of a TM domain , which might mitigate against the detection of positive selection in this protein domain . HIV-1 is well adapted to replicate in human cells , but fails to replicate in many nonhuman primate cells . This is in large part because it is unable to evade or antagonize the species-specific variants of antiviral genes , such as TRIM5 and APOBEC3 , which show evidence of positive selection that is assumed to have resulted from past retroviral epidemics [21]–[23] . Tetherin represents a third example of an antiviral gene in monkeys that exhibits activity against intact HIV-1 as a consequence of positive selection in the primate lineage . Thus an array of antiviral molecules limit the replication of primate lentiviruses in non-natural host cells , creating barriers to zoonosis , and revealing potential opportunities to mobilize intrinsic antiretroviral defenses by therapeutic inhibition of the activity of their viral antagonists .
A hu-tetherin cDNA , cloned into pCR3 . 1 vector , and its N-terminally HA tagged counterpart have been described previously [4] . An internally HA-tagged tetherin expression construct , pCR3 . 1/hu-tetherin-HA was derived from this by inserting an NheI restriction site at nucleotide position 463 of the tetherin gene . Thereafter , complimentary oligonucleotides encoding an HA epitope tag were inserted into the NheI site . Similarly , the tetherin coding sequence from rhesus macaque , African green monkey chimpanzee and mouse was amplified using cDNA generated from IFN-alpha treated 221 , COS-7 , chimpanzee fibroblasts , and NIH3T3 cells respectively . For the monkey tetherins , the HA epitope was inserted at a position orthologous to nucleotide 463 as described above , while a N-terminally tagged murine tetherin construct was used . Thereafter , overlap-extension PCR approaches were used to exchange TM domain segments between human and monkey proteins , or to introduce point mutations into the hu-tetherin sequence . 293T cells were maintained in DMEM media supplemented with 10% fetal calf serum and gentamycin , as were HeLa-TZM cells which express CD4 and CCR5 and contain a lacZ reporter gene under the control of an HIV-1 LTR . To measure tetherin and Vpu activity , 293T cells were seeded in a 24 well plate at a concentration of 1 . 5×105 cells/well and transfected the following day using polyethylenimine ( PolySciences ) with 500 ng of an unmanipulated HIV-1 proviral plasmid NL4-3 ( WT ) or a Vpu-defective counterpart NL4-3 ( delVpu ) . Additionally , 50 ng of a tetherin expression plasmid and 50 ng pCR3 . 1/cherry fluorescent protein ( to monitor transfection efficiency ) were included in the transfection . In experiments where the level of tetherin was varied , the tetherin expression plasmids were serially diluted from 200 ng to 12 . 5 ng per transfection and pCR3 . 1 was used as a DNA filler . Stable WT and mutant hu-tetherin-HA expressing 293T-derived cell lines were generated by retroviral transduction , as previously described [8] . Transfected 293T cells were place in fresh medium at 20 hrs post transfection and virion containing cell supernatants were harvested and filtered ( 0 . 2 µm ) at 40 hrs post transfection . Infectious virus release was determined by inoculating , in triplicate , sub-confluent monolayers of HeLa-TZM cells seeded in 48 well plates at 2 . 5×104 cells/well with 50 µl of serially diluted supernatants . At 48 hrs post infection , ß-galactosidase activity was determined using GalactoStar reagent as per the manufacturer's instructions . The remainder of the virion containing supernatant ( 450 µl ) was layered onto 800 µl of 20% sucrose in PBS and centrifuged at 20 , 000 g for 90 minutes at 4°C and virion yield determined by western blot assays Pelleted virions and the corresponding cell lysates were resuspended in SDS-PAGE loading buffer and separated on NuPAGE Novex 4–12% Bis-Tris Mini Gels ( Invitrogen ) . Proteins were blotted onto nitrocellulose membranes . Thereafter , HIV-1 Gag or capsid proteins , as well as tagged tetherin proteins were revealed using anti-capsid and anti-HA antibodies and chemiluminescent detection reagents , as described previously . 293T cells stably expressing either WT or mutant ( delGI/T45I ) tetherin-HA were plated on poly-D-lysine coated dishes ( Mattek ) . The following day the cells we infected with VSV-G pseudotyped HIV-1 ( WT ) or HIV-1 ( delVpu ) variants that carried Cerulean-FP ( CFP ) embedded in the stalk region of the matrix domain of Gag . The virus dose was chosen so that approximately 40% of the cells were infected . At 48 h after infection , cells were fixed but not permeabilized in order to confine tetherin-HA staining to surface expressed protein . Fixed cells were sequentially incubated with an anti-HA monoclonal antibody ( Covance ) followed by an anti mouse IgG Alexafluor 594 conjugate . The cells were imaged using a Deltavision microscopy suite and infected cells ( identified by the presence of CFP fluorescence ) were scored for the presence of intense tetherin-HA staining on the cell surface . The aforementioned human , chimpanzee , rhesus monkey , African green monkey and mouse sequences were included in an analysis for positive selection . In addition , tetherin genes were amplified from lymphocyte RNA from several rhesus macaques ( n = 8 ) pigtail macaques ( n = 1 ) , crab eating macaques ( n = 6 ) and sooty mangabeys ( n = 6 ) . These were cloned using a kit TOPO-TA kit ( Invitrogen ) and the sequences of multiple clones determined . Representative alleles were included in the analysis described below . Additional tetherin sequences from gorilla , gibbon , and marmoset were retrieved from the raw data archives of ongoing genome sequencing projects by TraceBLAST ( http://www . ncbi . nlm . nih . gov/BLAST/Blast . cgi ) , using each of the hu-tetherin coding exons as a separate query . Sequences were aligned using Macvector , and adjusted manually . Codon-based nucleotide alignments were used in conjunction with phylogenetic trees generated using the DNAPARS program ( PHYLIP ) as input for the random effects likelihood ( REL ) program ( HyPhy ) to detect positive selection . Input files for analysis using CODEML in the PAML suite ( version 3 . 14 ) were generated by first aligning amino-acid sequences using the CLUSTAL-W algorithm , converting the alignment back to nucleotides , and adjusting manually where necessary using MEGALIGN ( DNASTAR , Madison , WI ) . Tree files were generated by Neighbor-Joining , and sites with dN/dS>1 were identified using the resulting tree or a tree constrained to accept the known major branches of primate evolution as input , with similar results . The F3X4 model of codon frequencies was used for all analyses in CODEML . Paired , nested models of sequence evolution implemented in CODEML ( M0 , M3; M1 , M2; M7 , M8 ) were also compared using the likelihood ratio test . Evaluation with the chi-square test assumed either 4 degrees of freedom ( M0 , M3 ) or 2 degrees of freedom ( M1 , M2; M7 , M8 ) .
|
Tetherin is a cell surface protein that acts as an antiviral defense . It functions by tethering newly assembled HIV-1 particles to the surface of the infected cell , such that the viral particle is unable to depart and disseminate to other , uninfected cells . HIV-1 possesses an antagonist of tetherin , termed Vpu , that abolishes tetherin function . We found that HIV-1 is an effective antagonist of human and chimpanzee variants of tetherin but is unable to antagonize tetherins from two monkey species . Additionally , we found that sequence differences in a portion of the protein that is embedded in cell membranes determined whether or not it could be antagonized by Vpu . Since the Vpu protein is alsi a membrane embedded protein , this result suggests that Vpu and tetherin interact within cell membranes . We also show that tetherin has been evolving rapidly , and has likely been placed under selective pressure to change sequence . Notably , portions of tetherin that appear to have been placed under selective pressure coincide with positions that influence Vpu antagonism . Therefore , the evolutionary history of primates determines the effectiveness of HIV-1 Vpu in modern species . Thus , tetherin could impose a barrier to cross species transmission of retroviruses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"virology/immunodeficiency",
"viruses",
"virology/host",
"antiviral",
"responses"
] |
2009
|
Species-Specific Activity of HIV-1 Vpu and Positive Selection of Tetherin Transmembrane Domain Variants
|
The schistosome esophagus is divided into anterior and posterior compartments , each surrounded by a dense cluster of gland cell bodies , the source of distinct secretory vesicles discharged into the lumen to initiate the processing of ingested blood . Erythrocytes are lysed in the lumen , leucocytes are tethered and killed and platelets are eliminated . We know little about the proteins secreted from the two glands that mediate these biological processes . We have used subtractive RNA-Seq to characterise the complement of genes that are differentially expressed in a head preparation , compared to matched tissues from worm tails . The expression site of representative highlighted genes was then validated using whole munt in situ hybridisation ( WISH ) . Mapping of transcript reads to the S . mansoni genome assembly using Cufflinks identified ~90 genes that were differentially expressed >fourfold in the head preparation; ~50 novel transcripts were also identified by de novo assembly using Trinity . The largest subset ( 27 ) of secreted proteins was encoded by microexon genes ( MEGs ) , the most intense focus identified to date . Expression of three ( MEGs 12 , 16 , 17 ) was confirmed in the anterior gland and five ( MEGs 8 . 1 , 9 , 11 , 15 and 22 ) in the posterior gland . The other major subset comprised nine lysosomal hydrolases ( aspartyl proteases , phospholipases and palmitoyl thioesterase ) , again localised to the glands . A proportion of the MEG-encoded secretory proteins can be classified by their primary structure . We have suggested testable hypotheses about how they might function , in conjunction with the lysosomal hydrolases , to mediate the biological processes that occur in the esophagus lumen . Antibodies bind to the esophageal secretions in both permissive and self-curing hosts , suggesting that the proteins represent a novel panel of untested vaccine candidates . A second major task is to identify which of them can serve as immune targets .
Adult schistosome worms reside in the host vascular system actively feeding on blood that contains antibodies , complement factors and effector leucocytes , yet they are apparently unaffected by this ‘toxic’ diet . Indeed , their attested longevity in the hepatic portal system ( Schistosoma mansoni and S . japonicum ) or the venous plexuses around the bladder ( S . haematobium ) illustrates the sophisticated yet poorly understood mechanisms they must deploy to evade the host immune response in such a hostile environment [1] . The schistosome alimentary tract comprises an oral sucker around the mouth , a short esophagus and an extended gut caecum that runs to the extreme posterior [2] . The caecum comprises a syncytial gastrodermis that is both secretory and absorptive , and an associated network of muscle fibres responsible for peristalsis . It occupies a larger proportion of body cross section in females ( 16% ) than males ( 6% ) [3] , reflecting the disparate balance between nutrient uptake across the body surface and gut in the two sexes [2] . The proteolytic enzymes responsible for breakdown of ingested proteins in the acidic environment of the gut lumen have been well researched [reviewed in 2] . In addition , a proteomic analysis of the vomitus released by worms in short term culture [4] has revealed the presence of other hydrolases , as well as ‘transport’ proteins capable of binding lipids ( e . g . saposins ) and inorganic ions ( ferritin , calumenin ) . In vitro feeding experiments with labelled dextran have demonstrated the occurrence of endocytosis at the gastrodermal surface [4] , while laser capture microdissection has been used to identify genes encoding transporters putatively expressed on the luminal surface of the gastrodermis [5] . In contrast , the role of the esophagus has been under-appreciated and little researched since the first ultrastructural descriptions several decades ago [6 , 7] . However , we have recently shown that , instead of being just a conduit , it actually initiates the processing of ingested blood before it reaches the gut lumen [8] . The esophagus is divided into anterior and posterior compartments , each surrounded by an associated mass of cell bodies and lined by a syncytial layer of cytoplasm continuous with the surface tegument . The posterior mass was designated as a gland decades ago and we have recently shown in S . japonicum that the anterior cell mass is also a distinct secretory organ [9] . Both cell masses synthesise proteins for secretion into the lumen . Video recording of feeding [8] and in-vitro experiments with membrane-labelled erythrocytes [4] have revealed their lysis in the lumen; the label transfers primarily to the membranes of the posterior compartment . The two observations explain why intact erythrocytes are seldom seen in the lumen [8] . In contrast host leucocytes accumulate within the posterior lumen as a central plug around which incoming blood flows [8] . Furthermore , these tethered leucocytes are structurally damaged , as are those which reach the gut lumen . Despite intact platelets being observed in the anterior compartment [10] , ingested blood does not clot in the lumen , implying the existence of anticoagulant mechanisms . Collectively , these observations confirm the esophagus as a crucial site for interaction of host blood with parasite products . Specific expression of three microexon genes ( MEGs [11 , 12] ) , namely MEG-4 . 1 [13] , MEG-4 . 2 and MEG-14 [8] , and one venom-allergen-like ( VAL; [14] ) gene , VAL-7 [15] was revealed in the posterior esophageal gland of S . mansoni by whole mount in-situ hybridisation ( WISH ) . In addition , seven proteins ( six MEGs and VAL-7 ) have been localised to the posterior esophageal gland of S . japonicum by immunocytochemistry [8 , 16] . Furthermore , the demonstration of host IgG binding to the esophageal lumen of both mouse and hamster worms in vivo [8] raised the possibility that esophageal proteins might be targets of the host response . Most recently we have obtained evidence that rhesus macaques self-cure from an established S . japonicum infection by producing antibodies that target esophageal secreted proteins [16] . The functions of the esophagus are disrupted , leading to cessation of feeding , starvation and ultimately death of established worms [16 , 17] . Clearly , if we are to understand esophageal function better we need more information about the proteins secreted into the esophagus lumen that interact with incoming blood . We have also suggested such proteins represent an entirely new group of targets that might be exploited for vaccine development , due to their critical role in blood feeding and their accessibility to antibodies [16] . The advent of new and cheaper technologies has made comparative transcriptome analysis by direct sequencing feasible . We have used the massive parallel capacity of ion semiconductor sequencing on an Ion Torrent instrument to investigate differential gene expression in the esophageal region of adult male S . mansoni . Schistosomes possess epithelia ( tegument , gastrodermis ) and rudimentary organ systems ( muscles , nerves and sense organs , alimentary tract , protonephridial system , parenchyma ) present throughout the whole body but the solid acoelomate body plan means they are not readily isolated for analysis . However , the cell masses surrounding the anterior and posterior esophageal compartments , plus the paired cerebral ganglia of the nervous system are unique to the esophageal region . We therefore reasoned that a subtractive comparison of the patterns of gene expression in heads and tails would delineate this unique ‘head’ subset . We present here the results of that comparison , which highlighted a group of differentially expressed genes , many encoding secretory proteins , and we have validated the expression of representatives to the cell bodies of the anterior or posterior esophageal glands . These data lay the foundations for a deeper understanding of blood processing in the worm esophagus and provide a panel of proteins that can be screened for immunoreactivity against sera from permissive and self-curing hosts .
The procedures involving animals were carried out in accordance with the Brazilian legislation ( 11790/2008 ) . The protocol for maintenance of the S . mansoni life cycle was reviewed and approved by the local ethics committee on animal experimentation , Comissão de Ética no Uso de Animais ( CEUA ) , Universidade Federal de Ouro Preto ( UFOP ) , and received the protocol no . 2011/55 . Balb/c strain mice were infected with approximately 200 cercariae and adult worms obtained by portal perfusion of animals at 6–7 weeks later , using RPMI-1640 medium buffered with 10mM HEPES ( Sigma-Aldrich , St Louis , MO , USA ) . After extensive washing in the same medium and removal of tissue debris and any damaged individuals , parasites were fixed instantly by immersion in RNAlater ( Invitrogen , Paisley , UK ) . Approximately 400 male worms were individually viewed at x30 magnification under a dissecting microscope , carefully held with fine watchmakers forceps ( Ideal Tek , Chiasso , Switzerland ) and the head region detached along the line of the transverse gut using Vannas scissors ( John Weiss , Milton Keynes , UK ) . Two hundred tails , defined as the posterior third of the male body to exclude the testes , were similarly excised in order to obtain the same amount of biological material . Before extraction , the two sample pools were disrupted on ice using a tissue grinder until they appeared completely homogeneous . Total RNA was extracted using an RNeasy Micro kit ( Qiagen , Manchester , UK ) . Briefly , the homogenized lysate was centrifuged for 3 min at full speed to pellet the debris . The supernatant was transferred to a clean tube and mixed with 1 volume of 70% ethanol . The mixture was then transferred to an RNeasy MinElute spin column and centrifuged for 15s at ≥8000xg . After washing and DNA digestion with DNase I , total RNA was eluted with 10μl RNase free water . Messenger RNA was further purified from total RNA using a Dynabeads mRNA DIRECT kit ( Life Technologies , Warrington , UK ) according to the manufacturer’s instructions . In short , the Dynabeads Oligo ( dT ) 25 beads were washed in one well of a 96-well plate sitting on a magnetic stand . Total RNA was diluted as required and heated at 70°C for 2 min then mixed with an equal volume of Lysis/Binding Buffer . The denatured RNA mixture was transferred to the well containing the beads and incubated for 5 min to allow mRNA binding . After washing , the mRNA was eluted from beads with pre-warmed ( 80°C ) nuclease-free water . Two rounds of mRNA isolation were performed in the same well in order to achieve a high quality mRNA yield . All reagents and equipment used was obtained from Life Technologies unless otherwise stated . Libraries were prepared for RNA-sequencing using the Ion Total RNA-Seq Kit v2 , and the recommended protocol for whole transcriptome library preparation from <100 ng Poly ( A ) RNA . In brief , RNA was fragmented using RNaseIII , and Ion Adaptors ( Mix v2 ) ligated to fragmented RNA prior to reverse transcription . cDNA was then purified and amplified using Ion Xpress RNA-seq Barcoded primers , with separate barcodes used for each sample . The yield and size distribution of each amplified cDNA library was assessed using the Agilent High Sensitivity DNA kit with the Agilent 2100 Bioanalyzer . Libraries were then pooled at equimolar concentrations , and diluted to 20 pM in preparation for sequencing . Two independent rounds of Ion Torrent sequencing were performed on pooled libraries to allow comparison of technical replicates , and the data combined for downstream analysis . In accordance with the recommended protocols provided , sequencing template preparation was performed using the Ion OneTouch system in conjunction with the Ion PGM Template OT2 400 kit , where template positive Ion Sphere Particles ( ISPs ) were prepared and subsequently enriched . Sequencing was then performed on an Ion Personal Genome Machine System , using an Ion 318 Chip v2 with the Ion PGM Sequencing 400 Kit . The assumption that all the major organ systems and tissues would be present in roughly equal proportions in both the head and tail samples was tested by compiling lists of signature proteins for comparison . The genes encoding cytosolic proteins were taken from the proteomic analysis of the SWAP fraction of adult worms [23] . The parenchyma was represented by the genes encoding glycogen metabolism proteins , culled from the genome database; together with muscle this tissue is the principal site of such activity [24] . The muscle and cytoskeletal genes were taken from the list of proteins identified in the Tris and UTCS fractions of frozen/thawed adult worms [25] . Tegument and gut-secreted proteins were compiled from the respective proteomics studies [26 , 27 , 28 , 4] , supplemented by annexins and tetraspanins annotated in the genome , and saposins from an infection array experiment [29] . The lists of representative glycosyl transferases and nervous system genes were compiled by key-word searching of the data using appropriate terms ( glucosyl , galactosyl , fucosyl , xylosyl , mannosyl , transferase and neur , synap , acetylch , dopam , transmit , respectively ) followed by manual editing . Eleven targets were chosen from the subset of genes highly enriched in the head samples for independent validation of the site of expression , using whole mount in-situ hybridization ( WISH ) . These were an Aspartyl Protease , Beta 1 , 3-galactosyltransferase , a Phospholipase A2 and MEGs 8 . 1 , 9 , 11 , 12 , 15 , 16 , 17 and 22 , with VAL-7 as a positive control [15] and the sense sequence of VAL-7 as the negative control . The method was performed on whole adult male and female worms as described by Dillon et al . [13] . The worms were first fixed in Carnoy’s solution , then in MEMFA ( 0 . 1 M MOPS , 2 mM EGTA , 1 mM MgSO4 and 3 . 7% formaldehyde ) before storage in ethanol at -20°C until use . Briefly , for the protocol worms were warmed to room temperature and rehydrated by 2x5 min washes , the first in 75% ethanol/25% phosphate-buffered saline ( PBS; pH 7 . 4 ) containing 0 . 1% Tween 20 ( PBSAT ) and the second in 50% ethanol/PBSAT . They were then transferred to 100% PBSAT for 3x5 min washes . After rehydration , parasites were permeabilized in a 10 μg/ml solution of PCR-grade proteinase K ( Roche , Germany ) dissolved in PBSAT , and refixed with formaldehyde . For probe synthesis , sequences of interest ( S1 Table ) were manufactured by Biomatik ( Cambridge , Canada ) and cloned into the plasmid pBSK ( + ) . Antisense RNA probes were obtained in vitro incorporating DIG-labelled dUTP ( Roche , Germany ) with T7 or SP6 RNA polymerase ( Promega , USA ) . The permeabilized worms were then incubated at 60°C for 2h in hybridization buffer ( 50% formamide , 5 x SSC , 100 μg/ml heparin , 1x Denhardt’s solution , 0 . 1% Tween 20 , 0 . 1% CHAPS and 5 mM EDTA ) with 1 mg/ml total yeast RNA added to block non-specific hybridization . After this step , the solution was replaced with fresh ( pre-warmed ) total RNA/hybridization buffer containing 1 μg/ml of synthesized DIG-labelled probe and hybridization was performed at 60°C overnight . After several washes , parasites were incubated with alkaline phosphatase-conjugated anti-DIG Fab fragments ( Roche ) overnight at 4°C . After more washes , parasites incubated with BM-Purple substrate were observed for colour development and photographed using the Microscope Eye-Piece Camera ( Dino-Lite , Taiwan ) . The full IonTorrent dataset has been deposited on the NCBI SRA site ( http://www . ncbi . nlm . nih . gov/sra ) under the Study number SRP064960 . The S . mansoni heads sample was designated as SRS1120313 and the experiment as SRX1353319 . Heads run 1 and heads run 2 have the accession numbers SRR2722034 and SRR 2722095 , respectively . The S . mansoni tails sample was designated as SRS1120316 and the experiment as SRX1353321 . Tails run 1 and tails run 2 have the accession numbers SRR2722255 and SRR 2722455 , respectively . New or improved gene annotations deposited on the EMBL TPA site received accession numbers as follows: MEGs 26–31 & MEGs 10 . 2 , LN898187-LN898193; Aspartyl protease ( Smp_018800 ) , LN898196; Phospholipase ( Smp_031180 ) LN898197; Phospholipase ( Smp_031190 ) , LN898198; MEG-32 . 1 ( Smp_123100 ) , LN898194; MEG-32 . 2 ( Smp_123200 ) , LN898195 .
The Ion Total RNA-Seq Kit v2 requires a minimum of 1ng polyA purified mRNA for library construction . Using the appropriate kits , extraction of the head sample homogenate yielded 190ng of total RNA , from which 1 . 92 ng mRNA was recovered , adequate for library construction; mRNA recovery from the tails was not a limiting factor . Ion Torrent sequencing of the DNA fragments from the two technical replicates of head and tail mRNA extracts yielded between 0 . 86 and 1 . 8 million reads in the four runs ( S2 Table ) . Approximately 65% of these were mapped by the Tophat and Cufflinks programmes to predicted genes in version 5 of the S . mansoni genome , thereby providing identities and Smp gene annotations . Frequency distributions of RPKM values , depicting transcript abundance in heads and tails , were virtually superimposed ( S1A Fig ) . A plot of the RPKM values for the technical replicates of heads ( S1B Fig ) and tails ( S1C Fig ) revealed the high degree of reproducibility and linearity ( correlation coefficients 0 . 99 and 0 . 98 respectively ) in the sequencing data , with a dynamic expression range between four and five orders of magnitude . A total of 8856 genes was represented by one or more reads; this reduced to 5010 genes when those with trivial numbers of reads were eliminated ( RPKM <16 ) . Of these , 2583 were more highly expressed in the heads and 2427 in the tails . A scatter plot of transcript abundance against the difference in expression between heads or tails ( Fig 1 ) delineated subsets of 97 genes in the head ( 1 . 95% ) and 80 in the tail ( 1 . 6% ) that were displaced more than fourfold either side of the equivalence line x = 0 . Of these , 23 genes were expressed in the heads only and 10 in the tails only ( points lying along the 45° line in Fig 1 ) . The intensity of expression of the head subset was greater ( mean RPKM 3963 , median 98; S3A Table ) than that of the tail subset ( mean RPKM 1246 , median 84; S3B Table ) and there was also a massive bias in differential expression in the head subset . The ratio of mean RPKMs H/T of x166 compared with a value of only x6 . 2 for the T/H ratio in the tail subset ( S3A & S3B Table ) . A further five genes were excluded from the heads analysis and 12 from the tails because there were five or fewer reads in either replicate , making a total of 92 and 67 genes for detailed analysis of expression , respectively . The subtractive RNA-Seq approach requires that the major schistosome tissues are equally represented in both head and tail preparations if it is to identify genes uniquely or predominantly expressed in esophageal structures . We tested this proposition by comparing the relative expression of genes encoding signature proteins ( S1D Fig; S4 Table ) . Individual paired RPKM values from the two technical replicates , displayed as a scatter plot , revealed the similarity in gene expression level for heads and tails , with a correlation coefficient of 0 . 951 . The diffuse schistosome nervous system ( NS ) ramifies through all tissues of the worm body but nerve cells are sparse and the level of expression of NS genes was the lowest for any signature tissue . The mean RPKM score of NS abundance in heads ( = 69 ) thus provides a benchmark for comparison of the levels of transcript abundance of other tissue signatures in the heads . The tegument and gastrodermis are the two principal interfaces with the external environment , where considerable biosynthesis of proteins for export takes place . The mean expression level of genes encoding tegument surface proteins ranged from 7 . 3 to 10 . 5 times the NS , apart from those encoding the transporter-linked ATPases at 1 . 3 times the NS ( Fig 2A ) . The mean relative transcript abundance for proteins secreted by the gastrodermis ranged from 10 to 17 times the NS , indicating a slightly higher level of biosynthetic activity than the tegument cell bodies; genes encoding the extended group of ten saposins , likely involved in lipid binding and transport , were the most active . Muscle and parenchyma are the most abundant tissues in the male schistosome body , with cytoskeletal proteins showing a mean expression level ( 13 . 5x NS ) similar to the gastrodermis , while the genes encoding cytosolic proteins ( e . g . glycolytic enzymes , chaperones ) were the most highly expressed of all signature proteins ( mean 74x NS ) . Surprisingly , the genes responsible for glycogen metabolism , indicative of parenchyma , were expressed at only 1 . 8 times the NS level . The glycosyl transferases , involved in the N or O-linked glycosylation of proteins destined for export , were expressed overall in the same range as the NS genes . We next examined the bias in gene expression by comparing the ratio of RPKM scores for each signature gene in heads versus tails ( Fig 2B ) . Despite the cerebral nerve ganglia being present in the head preparation , there was only a slight bias ( x1 . 1 ) towards the heads among the 80 genes scrutinised . The other signature groups showed a bias towards greater expression in the tails with ratios ranging from 0 . 7 to 0 . 93 , apart from the glycosyl transferases . Expression of these was skewed ( x1 . 36 ) towards the head preparation and suggestive of specialised glycan production in the region . Finally , scatter plots of the signature groups confirmed that the vast majority of the ~250 signature genes lay within a very narrow range either side of the equivalence line ( Fig 3 ) . This allowed us to set a generous margin of fourfold difference either side of the line to define differential expression , and thereby to pinpoint any outliers . On that basis not one of the 80 nervous system genes was differentially expressed ( Fig 3A ) , nor were any muscle and cytoskeleton ( Fig 3D ) or parenchyma and cytosol genes ( Fig 3E ) . However , four potential tegument genes from the extended family lists of annexins and tetraspanins were heavily skewed in expression ( Fig 3B ) ; these have not been identified previously by proteomic analysis of the tegument and one of them ( Smp_155580 ) was particularly abundant in the heads . There was one exception in the bias of the saposin genes towards greater expression in the tails ( Fig 3C ) , with Smp_028840 showing eightfold expression in heads , albeit at low intensity . Among the glycosyl transferases three genes stood out with a much greater bias towards expression in the heads ( Fig 3F ) , Smp_159490 , Smp_144260 and Smp_151220 being expressed at 16 , 43 and 214 times the level in the tails , respectively . With the eight exceptions noted above , the discrete group of 92 genes differentially expressed in the heads at more than four times the level in tails did not encode signature proteins . Overall transcript abundance was 57 times that of nervous tissue ( median 5x higher ) with one fifth of the group exclusively expressed in the heads ( S3A Table ) . The remarkable feature of this list , sorted by abundance ( S3C Table ) , was the identity of the top 20 genes . MEGs accounted for more than half the total , including the three already known to be expressed in the esophageal gland ( MEGs 4 . 1 , 4 . 2 and 14 ) , plus a further eight ( MEGs 8 . 1 , 8 . 2 , 9 , 11 , 12 , 15 , 16 & 17 ) , whose site of expression was not previously recorded ( Fig 4 ) . For this top-20 group , the mean RPKM value for abundance ( 18638 ) was 270x the NS level and the expression ratio of heads to tails was x604 . Indeed MEG-4 . 2 , the most highly expressed gene in the group , had an RPKM value of 66834 , x969 the mean level of signature genes in nervous system tissues that are located adjacent to the gland . A significant number of reads was not mapped to the genome by Tophat and Cufflinks so we performed a de novo assembly of all reads using Trinity . We examined this de novo assembly for the presence of transposons by BLAST searching with a file of 34 well-characterised elements . A total of 27 was identified , amounting to not less than 0 . 9% of head and 0 . 6% of tails reads , counting only the top hit . None showed a marked differential expression between head and tail samples , but the three most highly expressed were Saci-1 , -2 and -3 , previously described as displaying high transcriptional activity [30] . We then interrogated the Trinity data to identify novel differentially expressed protein coding genes not annotated in the genome . We first confirmed that there was no major discrepancy between the results of Tophat/Cufflinks mapping and Trinity de novo contigs by plotting the abundance and differential expression of 12 known MEGs identified by both methods . The correlation coefficient r for comparisons of their RPKM values was 0 . 88 and for the ratio of heads/tails was 0 . 93 , confirming a strong positive association between the two methods . The Trinity contigs were annotated by BLAST against v5 of predicted genes in the S . mansoni genome to obtain identities of known genes . The same cut-offs for abundance and difference were then applied as for Tophat/Cufflinks mapping and a total of 51 unannotated contigs was filtered out for manual analysis . In this group , the mean RPKM value for abundance ( 418; 6x the NS ) , was much lower than for the 92 differentially expressed genes detected by Cufflinks ( = 3963 ) . However , this might be anticipated , given that these unannotated contigs encode potential genes not detected by conventional methods . The mean expression ratio of heads to tails ( x62 ) is still very biased towards the heads and 63% were expressed in the heads alone . Sixteen of the novel genes were identified as encoding unannotated MEGs ( Fig 4 , S5 Table ) . These included six previously described ( MEGs 8 . 3 , 8 . 4 , 19 , 20 , 22 , 24 [31] , and a further eight that were entirely new , namely MEGs 26–31 plus two new members of existing families , MEGs 10 . 2 and 4 . 3 . Gene annotations for seven of these new MEGs were deposited on EMBL and received accession numbers LN898187-LN898193 ( MEG-4 . 3 was a partial sequence in a genome fragment ) . Finally hundreds of reads were found in the Trinity assembly for a previously annotated gene , MEG-3 . 4 , which has been removed from the genome database by the curators at GeneDB and was therefore not detected by Tophat/Cufflinks mapping . We extended the evaluation of MEG expression to the unfiltered Cufflinks and Trinity datasets to locate any MEGs that were not differentially expressed ( Fig 4 ) identifying five ( 1 , 5 , 6 , 13 and 24 ) in this way ( S3D Table ) . MEG-6 transcripts were particularly abundant ( >3000x the level in the NS ) but only 3 . 7x higher in heads than tails . The other four all had a low or moderate level of expression with MEG-5 , previously identified in tegument preparations by proteomics , having an RPKM of 1910 in heads and 832 in tails . One final MEG ( MEG-10 . 1 ) appears to be an outlier , expressed in tails only and possibly confined to a very scarce tissue . Analysis of gene structure of the novel MEGs described here ( Fig 5 ) reveals that all of them display the typical characteristics: two long 3' and 5' flanking exons and a central coding portion mostly composed of microexons ( <36bp ) . The portion coding for a signal peptide is mostly or entirely contained in the 5' long exon . Twenty-six of the thirty-three microexons ( 79% ) that encode MEGs 26–32 are symmetrical ( i . e . have a length divisible by 3 ) , which indicates an evolutionary pressure to favour alternate splicing without disruption of the open reading frame . As with most MEGs , those enriched in the head preparation tend to encode relatively small proteins ( average MW ~10 kDa; median MW ~7kDa ) . Analysis of all the MEG structures using the IUPred program reveals that 13 out of 27 predicted protein products enriched in the head preparation display more than 40% of their length as intrinsically disordered ( Fig 6A; S5B Table ) . These disordered regions are rich in threonine , serine and proline ( TSP ) residues ( S5C Table ) . Unsurprisingly , nine of these thirteen proteins ( MEGs 4 . 1 , 8 . 1 , 8 . 2 , 14 , 15 , 19 , 20 , 29 and 32 . 1 ) are predicted to be heavily O-glycosylated , with an average of 16 . 4 sites , 100% of them being located in the putative disordered regions ( Fig 6A ) . A further five MEGs ( 8 . 3 , 10 . 2 , 22 , 32 . 1 , 32 . 2 ) are predicted to be O-glycosylated proteins but without extensive regions of disorder ( S5B Table ) . In the MEG-8 family members there is a hydrophobic C terminus encoded by the long 5’ flanking exon ( Fig 6A ) . Clustal comparisons reveal that it contains conserved protein domains with characteristic signatures for each family member across the three schistosome species for which sequence data is available ( Fig 6B ) . This cross-species conservation reveals that the MEG-8 diversity is ancient , arising before speciation of the Genus Schistosoma occurred . In addition , MEG-15 also displays a relatively hydrophobic C-terminus . Another group of MEGs ( 9 , 12 , 26 , 27 and 28 ) preferentially expressed in the head , encode a small peptide that is predicted by Heliquest to contain an amphipathic helix with a hydrophobic interaction face ( Fig 6C and 6D ) . That still leaves approximately one third of the proteins encoded by esophageal MEGs that have no distinguishing features to provide a clue to putative function , other than a signal peptide . A second prominent group of genes detected by the Tophat/Cufflinks mapping ( and confirmed by the Trinity assembly ) were nine hydrolases ( Fig 7; S3D Table ) . Homology searching of NCBI nr indicates that these are likely to be of lysosomal origin . The group of six proteases annotated as subfamily A1A unassigned peptidases ( A01 family ) located on Chromosome 3 , were exclusive to the head preparation , and ranged in transcript abundance from 0 . 4 to 28 times the NS level . BLAST and Clustal searching revealed they encoded closely related aspartyl proteases , referred to as Cathepsin D homologues . Transcripts for two other hydrolases , annotated as Phospholipase A2 , were abundant ( 3 . 6x and 9x NS , respectively ) and almost exclusively expressed in the heads . The final hydrolase ( 20x NS ) , palmitoyl protein thioesterase 1 enzyme , removes thioester-linked fatty acyl groups from modified cysteine residues in proteins or peptides . The N terminal sequence of aspartyl protease , Smp_018800 , was extended by Clustal mapping ( S6A Table ) and the updated gene annotation deposited at EMBL under accession number LN898196 . The presence of a signal peptide on this and four other group members ( Smps 132470 , 132480 , 136830 and 205390 ) was confirmed by SignalP . ( The sixth protease , Smp_136720 , is an incomplete gene model lacking the 5’ end . ) The five proteases also contained one to three copies of the consensus N-X-S/T sequence , indicating suitable sites for N-linked glycosylation . Palmitoyl thioesterase possessed a signal peptide plus N glycosylation sites and we were able to improve the gene models for the two Phospholipases using Trinity assemblies ( S6B & S6C Table ) , to reveal the presence of signal peptides and N glycosylation sites in both . These new gene annotations were deposited at EMBL under accession number LN898197 and LN898198 . All the evidence indicates that the nine hydrolases are destined for the lysosomal pathway and will have optimal enzymatic activity at an acid pH . As VAL-7 was already known to be expressed in the esophageal gland , and is a member of a large family of secreted proteins potentially important in modifying host responses , we searched our datasets for other VALs . Only VAL-7 was prominent and provides an internal control for the subtraction approach . The full length CDS for VAL-7 deposited at GenBank is divided without overlap between two genome scaffolds with two Smp designations . Their respective scores for abundance and difference are close on the scatter plot ( Fig 7 ) providing a strong indicator that the subtractive RNASeq method can produce reliable results . Only one further gene for VAL-13 , came above the >4-fold difference threshold with a score 2 . 1x the NS level , compared with a mean of 223x for VAL-7 . Five other VALs ( 16 , 8 , 12 , 11 and 6 ) were more evenly distributed between heads and tails ( 5B ) and it is notable that three of them ( 6 , 11 , 16 ) belong to Group 2 , lacking a signal peptide . The remaining 37 annotated genes mapped by Tophat/Cufflinks fell into two broad groups , 14 that showed a moderate level of differential expression and abundance , and a more compact group of 23 with a low differential; they were classified by putative function ( S3E Table ) . A cytosolic calmodulin-like calcium binding protein ( Smp_096390 ) was the most abundant transcript . A group potentially most relevant to esophageal secretion , and probably located along the secretory pathway , included genes encoding transmembrane emp24 domain containing protein 7 ( Smp_140180 ) involved in vesicular protein trafficking , transmembrane protein 63A ( Smp_143750 ) inserted in the membrane of lysosomes , GPI ethanolamine phosphate transferase 2 ( Smp_155490 ) involved in GPI-anchor formation and Longevity-assurance gene 1 ( LAG 1; Smp_122050 ) that facilitates transfer of GPI-anchored proteins from the endoplasmic reticulum to the Golgi apparatus . Finally , three putative nervous system transcripts at characteristic low abundance , and not in the list of signature NS genes were of note . Neuropeptide F prepropeptide ( Smp_088360 ) , tryptophan hydroxylase ( Smp_174920 ) and Catechol-o-methyltransferase ( Smp_198020 ) could represent markers for the cerebral ganglia , although the most skewed is only 16x the level in tails . The members of the largest grouping ( one third ) within the differentially expressed subset were annotated as hypothetical proteins , lacking homology to anything outside the Trematoda . Searching of the longest open reading frame for individual genes , in part hampered by incomplete gene models , yielded few that encoded signal sequences or transmembrane domains , a point dealt with in the Discussion . However , utilising a combination of Trinity assembly data and searching of publicly available EST databases , we were able to extend the models for two genes with abundant and differential expression in the heads , Smp_123100 and Smp_123200 , situated on chromosome 6 . Moreover , mapping of the exons to the chromosome revealed that both had a central region encoded by microexons; due to their homology we designated them MEG-32 . 1 and MEG-32 . 2 , respectively ( Fig 5 ) . Improved gene annotations for these two MEGs were deposited on EMBL and received accession numbers LN898194 and LN898195 . The two proteins are predicted to be membrane-anchored at both N and C termini to form a threonine-rich hairpin loop that is O-glycosylated . This makes a total of 12 previously annotated , 13 novel , and two reassigned MEG genes identified in the head preparation in the present study . Although not the focus of our study , we also analysed the markedly different set of 72 genes expressed more than fourfold higher in the tails than heads ( S3F Table ) . The largest group of 19 were annotated as encoding hypothetical proteins , primarily with homologs only among other Trematoda . Two of these ( Smp_177580 , Smp_201270 ) were the most abundant differential transcripts in the tails ( 771x and 362x the NS level ) . The second largest group encoded proteins of the extracellular matrix , and adhesion molecules such as protocadherin involved in cell attachment . A collagen ( Smp_135560 ) and a dynein light chain were both abundant ( 26x and 19x the NS level ) and among the most differentially expressed ( 8 . 5x and 52x the level in heads , respectively ) . Seven genes putatively associated with the gastrodermis , including two cathepsins and a saposin , could indicate some regional specialisation of the gut . The group of genes encoding four female-specific proteins in the posterior half of the male worm seems incongruous since they are involved in egg shell formation and predicted to be expressed in vitelline follicles but such follicles with their associated mRNA have been detected in male worms [13]; the most abundant transcript was present at 14x the NS level . The remaining annotated genes all with low levels of expression , encoded proteins involved in signalling pathways ( 6 ) , nuclear function ( 4 ) and miscellaneous processes ( 12 ) . Expression of the gene for MEG-10 . 1 in the tails at 6 . 3x the NS level was noted above . Detection of gene expression using WISH was successful for all 12 selected targets in males and eight in females ( Fig 8 ) . At low magnification the specificity of target gene expression only in the worm anterior between oral and ventral suckers is confirmed ( S2 Fig ) . At higher magnification , four of the genes , MEGs 12 , 16 , 17 and Phospholipase A2 were revealed as exclusively expressed in the mass of cells surrounding the anterior esophageal compartment , confirming its status as a distinct gland in S . mansoni . These are the first identified genes expressed in this region . Expression of the remainder , together with the VAL-7 positive control , was confined to the posterior esophageal gland cell bodies . They comprised two hydrolases ( aspartyl protease and palmitoyl thioesterase ) , five MEGs ( 8 . 2 , 9 , 11 , 15 and 22 ) plus a glycosyl transferase ( β1 , 3-galactosyltransferase ) . Expression of the five MEGs plus aspartyl protease was also detected in the posterior esophageal gland of female worms , whereas only MEG-12 expression was detected in the female anterior esophageal gland . The time for colour development after addition of substrate , and to a lesser extent the intensity of the signal corroborate the estimate of mRNA abundance represented by the RPKM score ( S3 Fig ) . The WISH targets with a high log2 RPKM between 13 . 3 and 16 all developed within 1–2 hours . The remaining six targets divide into two groups with medium ( 3–5 hrs ) and slow ( 6–10 hrs ) development time . The slow developers have log2 RPKMs between 10 . 4 and 12 . 7 , the medium developers between 8 and 10 . The confounding factor is that the length of probe , containing dig-labelled bases to which the detection antibody attaches , was twice as long in the medium as the slow developers . This illustrates the complexity of the WISH protocol and the difficulties for quantitation .
Our aim in this study was to obtain an insight into those genes expressed in the distinctive tissues of the schistosome esophagus that encode the proteins involved in the initial processing of ingested blood . The difficulties in characterising patterns of gene expression that occur in the discrete organ systems of an acoelomate metazoan with a solid body plan should not be underestimated . Laser capture microdissection [5 , 32 , 33] has been applied but the amount of tissue obtained and the precision needed to excise the organ of choice without contamination , are major limitations . Moreover , the studies to date have used microarray analysis to detect differences in gene expression between tissues , a technique which has inherent limitations . The fixed design of the array , especially if coverage is partial [32] , leaves gaps in the repertoire and furthermore does not permit new genes to be identified . The dynamic range of detection is also limited ( typically a maximum of 200-fold ) , due to high background levels , cross hybridisation and saturation of signals [34] . Rapid advances in technology have quickly led to the adoption of RNA-Seq as the method of choice to characterise transcriptomes from many sources [34] . The existence of a well-annotated gene assembly for S . mansoni [11 , 35] is a singular advantage and RNA-Seq can also identify novel coding sequences . The technology has a very low ( if any ) background , no upper limit for quantification and a dynamic range of 4–5 orders of magnitude [34] . Our RPKM scores of expression ranging from ~2 x 100 to 7 . 2 x 104 for heads and ~2 x 100 to 5 . 3 x 104 for tails were of that order . It has been estimated , using stringent criteria , that four million mapped reads of ~35 bases provided 80% coverage of gene expression in yeast [34] . We achieved 1 . 65 and 1 . 9 million mapped reads of mean length 117 bases , for heads and tails respectively , which detected 8856 genes or 82% of the predicted total . As we were not seeking an overview of the complete transcriptome , we deliberately excluded from analysis genes with fewer than five detected transcripts in both samples . This still left ~ 5000 genes , representing 42% coverage , to be evaluated . In comparison , the first qualitative , genome-wide analysis of S . mansoni [36] was performed on ~125 , 000 sequences generated largely from mini-libraries by the ORESTES protocol [37] . Our subtractive RNA-Seq approach would be equally applicable to characterise differential gene expression in other adult worm tissues such as the testis , ovary , uterus and ootype . The core of our strategy was the isolation by microdissection of the entire esophageal region and matching tails from adult male bodies stabilised with RNALater . We then generated the transcript datasets and used the subtractive approach , based on the dual criteria of abundance and differential expression , to delineate the set of genes exclusively or predominantly expressed in the head preparation . Surprisingly , nervous system genes were not prominent in the heads despite the presence of the cerebral ganglia and we must attribute this to the diffuse nature of the schistosome nervous system throughout the whole body . This left us with the proposition that the ~90 differentially expressed genes mapped by Tophat/Cufflinks , plus the novel genes detected by the Trinity assembly , were expressed in the cell bodies of the two esophageal glands . ( Note that there is no protein synthetic machinery in the syncytial lining of the esophagus . ) That many of the highlighted gene models are partial and that we found new genes in the most intensively studied schistosome species can be explained by two factors . First , the esophageal glands comprise only a tiny fraction of the worm body so their transcripts will be severely under-represented in the whole worm homogenates used hitherto as a source of mRNA , depriving programmes like Evidence Modeller [38] of the resource they need for gene annotation . The second is that de novo gene finding programmes look for patterns of bases not occurring by chance , thus excluding short runs that comprise the microexons of MEGs . We compared transcript distribution between heads and tails for signature genes , primarily identified by our previous proteomic studies , to determine whether the major schistosome tissues were equally represented; our data amply confirm this supposition . No outliers were detected in the lists of signature proteins from the cytoskeleton , and cytosol , underlining the ubiquity of muscle and parenchyma in both head and tail preparations . Similarly , no known tegument markers or genes encoding constituent of worm vomitus originating in the gastrodermis were differentially distributed . However , two annexins and two tetraspanins were highly biased . BLAST searching with the most abundant annexin , Smp_155580 , indicates that the N-terminus of this protein is missing so no conclusion is possible about whether it is a candidate for release into the esophageal lumen . Similarly , a putative saposin , Smp_028840 , was highlighted as possible candidate for esophageal secretion . Unfortunately , evidence for a saposin domain is weak ( Prosite & NCBI CDD searches ) and the sequence lacks a signal peptide; this gene was also detected as differentially expressed by a laser capture study [33] . The distribution of glycosyl transferase expression was investigated because bioinformatic analysis of MEGs 4 . 1 and 14 indicated that they were O-glycosylated [8] , and the presence of O-glycans in the posterior esophageal gland had been demonstrated by lectin staining [8 , 39] . The three differentially expressed transferases , one of them validated by WISH in the posterior gland , raise the possibility that some proteins exported from the esophageal glands are decorated with novel glycan structures not found in other worm tissues , an observation that could have immunological consequences . The importance of the secretory pathway in the esophageal cell masses is also underlined by the enrichment of transcripts from five genes involved in the intracellular vesicle transport pathway . A major finding of this study was the marked expression of MEGs in the head preparation , both in term of transcript abundance and differential . A total of 27 transcripts from 22 out of 32 MEG families ( two-thirds ) was detected in the schistosome head region , making it the most intense site for the expression of this enigmatic group of genes so far discovered . Furthermore , combining the results of this and our previous studies [8 , 16] using WISH and immunocytochemistry we can now be confident that three MEGs ( 12 , 16 and 17 ) are expressed in the anterior gland and nine ( 4 . 1 , 4 . 2 , 8 . 1 , 8 . 2 , 9 , 11 , 14 , 15 and 22 ) in the posterior gland . The second and complementary observation was the marked differential expression of nine genes encoding lysosomal hydrolases in the head region , with phospholipase A validated by WISH to the anterior gland , and aspartyl protease and palmitoyl thioesterase to the posterior gland . The mean RPKM scores for the four genes we have shown are expressed in the anterior esophagus and the 12 in the posterior esophagus [8 , 13 , 15 , 16] are 6053 and 24767 respectively . The posterior gland is approximately 2 . 5 times the volume of the anterior [8] suggesting a roughly equal transcriptional activity on a tissue mass basis . However , the RPKM scores are 10 to 20 times the mean values for tegument cell bodies and gastrodermal epithelium ( 598 and 1192 , respectively ) . We conclude that the glands are indeed a hotspot for gene transcription in the male worm body , potentially of secretory proteins destined for export into the esophagus . Our recent research has provided ultrastructural evidence for the secretion of vesicle contents from both anterior and posterior glands , into the esophageal lumen [8 , 9] . We also noted that the morphology of the ‘light vesicles’ in the anterior gland was akin to that of primary lysosomes , raising the possibility that lysosomal enzymes were secreted into the esophagus lumen [9] . Such lysosomal secretion is a well-established feature of the gastrodermis [4] . Our immunocytochemical observations provide direct evidence for the secretion of five MEG-encoded proteins and VAL-7 into the esophagus lumen [16] . Moreover , we have now identified two MEG proteins ( 8 . 2 and 15 ) and two lysosomal hydrolases ( aspartyl protease Smp_136830 & palmitoyl thioesterase ) in worm vomitus preparations ( WCB & LXN , personal communication ) . In terms of RPKM score , these identities are #s 2 & 3 on the MEG list and #s1 and 2 on the hydrolase list , illustrating the relative sensitivity of RNA-Seq versus proteomic detection . The above observations make a strong case that the protein products of the differentially expressed microexon and hydrolase genes identified in this study , are secreted into the esophagus lumen to interact with ingested blood . This poses the question as to their role in the esophageal processes that we have delineated [4 , 8] . These include erythrocyte lysis , leucocyte tethering and killing , disposal of platelets and prevention of clot formation . It should not be forgotten that there are other unannotated genes among the ~140 differentially expressed transcripts that may have a role in these processes . Predicting functions and devising assays for proteins with little or no homology to anything outside the Genus Schistosoma is a daunting task . However , we can make some inferences from predicted primary and secondary structures . Several of the MEG-4 , -8 and -15 families ( eight of the esophageal MEGs ) possess a central TSP-rich , intrinsically disordered region predicted to be heavily O-glycosylated . We have already demonstrated that the glycosylation of SmMEG-4 . 1 causes a gel shift of more than 70 kDa above the MW predicted for the protein alone [8] . We previously showed that the MEG-4 family proteins possessed a conserved C-terminus between schistosome species [8] . We have now shown that the MEG-8 family members all display a hydrophobic region at their C terminus , each again possessing a motif highly conserved between species . In MEG-4 we suggested that this region might target host leucocytes , e . g . by binding via its C terminus to a pan-leucocyte marker such as CD45 [8] . The same is true for the MEG-8 motifs and equally , plasma proteins or leucocyte secretions could be the intended ligands . An alternative possibility is that these O-glycosylated proteins might use the C-terminal motifs to organize themselves in a similar way to secreted mucins . There , unstructured , heavily glycosylated chains ( the TSP regions ) are connected by the interaction of hydrophobic domains , creating a net that confers to the mucus a gel-like consistency with viscoelastic proprieties [39] . The immuno-localisation of both SjMEG-4 . 1 [8] and SjMEG-8 . 2 [16] in a cocoon-like association with tethered leukocytes in the lumen of the S . japonicum esophagus provides visual evidence for a mucus-like complex that traps incoming leukocytes . TEM observations have revealed that the parallel arrays of material ( 0 . 08 x 0 . 1μm ) contained in the crystalloid vesicles of the posterior esophageal gland are released intact to the esophageal lumen , [8] and may cluster into larger aggregates , indicating the self-affinity/assembly of some molecular constituents . The largest aggregate measured to date ( 1 . 0μm x 0 . 6 μm ) in an electron micrograph represents a ~50-fold accretion . Moreover , the 40% greater repeating unit in the aggregates , compared to the vesicles [8] , indicates an expansion of the electron lucent layers after release , consistent with the swelling of O-glycans . It is thus plausible that the parasite adopts a “capture and defuse” strategy in the esophagus lumen whereby leukocytes are quickly immobilised and isolated by a mucus network , so preventing the diffusion of antibodies and defence proteins away from them . In contrast , those MEGs anchored by a transmembrane helix ( e . g . 14 , 29 , 32 . 1 , 32 . 2 ) , and also predicted to be O-glycosylated , are similar to the cell-associated mucins in humans in lacking the hydrophobic C-terminal region that might facilitate aggregation [40] . The most likely role for these membrane-anchored MEGs is to provide a protective lining coat of O-glycans for the entire esophagus . This suggestion is corroborated by the detection of a thin layer of neutral muco-substance lining the esophagus of the related blood fluke Schistosomatium douthitti [41] . The third obvious category of esophageal MEGs comprises the small peptides ( MEGs 9 , 12 , 26 , 27 and 28 ) that exhibit amphipathic helicoidal regions , with a hydrophobic face . Such peptides are widespread in animals and have both anti-microbial and haemolytic properties ( e . g . [42 , 43] ) . In the context of the schistosome esophagus they are capable of interacting with incoming erythrocytes and leucocytes to destabilize their membranes . Our recent observations on the localisation of S . japonicum MEG-9 confirm its association with the surface of leucocytes in situ in the esophagus lumen [16] . The lysosomal hydrolases we identified all have orthologues in mammals with well characterised properties that provide pointers to their potential function in blood processing . The putative secretion of two phospholipases , at least one of them from the anterior gland , strongly suggests a role in the lysis of erythrocytes as these cells pass through the two compartments . The transfer of the lipophilic dye PKH2 from labelled erythrocytes starts in the anterior compartment and is completed in the posterior [4] . TEM and confocal microscope images of intact erythrocytes in the anterior compartment and only a few ghosts in the posterior [44 , 8] are consistent with these observations . The palmitoyl thioesterase may also participate in the process of erythrocyte lysis via its ability to cleave the lipid anchor from proteins on the cytoplasmic face of plasma membranes . One such palmitoylated protein , p55/MPP1 [45] , found in the erythrocyte is an important component of the ternary complex that attaches the spectrin-based skeleton to the plasma membrane [46] . Thus we can envisage a cascade where short amphipathic MEG peptides such as MEG-9 or MEG-12 bind to and destabilise the erythrocyte membrane , enhancing the interaction of the two phospholipases with their plasma membrane substrates . Increased permeability then permits the palmitoyl thioesterase to enter and disrupt the cytoskeleton; the erythrocyte loses shape , leaks haemoglobin and is destroyed . Judging from our videos of worm feeding [8] the whole process takes only seconds . The secretion of six aspartyl proteases , at least one from the posterior compartment , indicates a powerful attack is also made on proteins in the plasma or on the external surface of host blood cells . Note that these enzymes are distinct from the one already described for the worm gut ( Smp_013040 ) [47] . The number of homologs suggests either redundancy of function , potentially as a means of immune evasion , or the existence of subtly different substrate specificities in the target proteins . The most obvious candidates are the components of the clotting cascade since clot formation does not occur in the worm esophagus . However , we have now detected fibrin localised in oval deposits in the anterior esophagus [16] , some of it coincident with host antibody , which may contribute to block secretion . A role for the aspartyl proteases in preventing this would require one or more to be synthesised by the anterior gland cells . A second potential function for these proteases could be the destruction of defence proteins released from the leucocytes that are , as revealed by TEM [8] , trapped and ultimately destroyed in the posterior lumen . One obvious corollary of the secretion of these lysosomal hydrolases is that they function at an acid pH optimum . The lumen of the schistosome gut has long been known to have a pH of ~4 . 5 [48] and now it appears that the process of acidification may begin in the esophagus . Nothing is known about the mechanism in schistosomes but in lower animals acidification of both the lysosomal interior and transepithelial compartments is effected by V-ATPases [49] . The transcripts of the genes encoding the complex of ~8 proteins that comprise this pump were all detected in our dataset , but were not differentially expressed—unsurprising given that both gut and esophagus may be acidified by the same process . If acidification of the esophagus lumen is confirmed , it would imply that the co-secreted MEG proteins also operate at an acidic pH . MEG hotspots described in our previous work were the head gland of the migrating schistosomulum ( MEG-3 family ) and the subshell envelope of the mature egg ( MEG-2 and MEG-3 family ) [12 , 50] . It has been suggested that the role of these egg- and larval-secreted MEG proteins is to interact with and modify vascular endothelial function [50] . In parenthesis , the only representative of the MEG-2 and 3 families found in the esophageal transcriptome was MEG-3 . 4 , not identified in the egg or larval secretions . It is plausible that the group of five non-differentially expressed MEGs ( 1 , 5 , 6 , 13 , 24 ) comprise a tegument hotspot since one of their number , MEG-5 , was previously detected in tegument fractions by proteomics [12] . The identification of a major hotspot of MEG expression in the worm esophagus , together with the expression of a group of lysosomal hydrolases , confirms the complexity of function that we have previously highlighted [8 , 9] . We have also observed binding of host IgG to the esophageal lumen , first in S . mansoni worms from permissive mice and hamsters [8] and more recently in S . japonicum worms from rhesus macaques undergoing self-cure [16] . In this last host the antibodies appear to target structures in both anterior and posterior compartments to alter morphology and disrupt function , potentially causing worm starvation and death [16] . Although we have suggested that the alternative splicing of MEGs generates a heterogeneous mixture of proteins that serve to confuse the immune system [12] , it appears that such a ploy can be circumvented by a host like the rhesus macaque . Collectively , the esophageal secretions that we have identified provide a novel and untested panel of vaccine candidates . With many available targets , the task is to discover the worm’s Achilles heel .
|
Schistosomes feed on blood and we have previously shown that its processing begins in the esophagus , which does not act simply as a conduit . It comprises anterior and posterior compartments , each surrounded by glands that secrete proteins into the lumen . Erythrocytes are ruptured as they pass through the compartments and leucocytes are tethered and killed but blood fails to clot . We wanted to identify the proteins secreted from these glands by sequencing the transcriptomes of head and tail preparations to pinpoint those messenger RNAs predominantly or exclusively present only in the heads . We found approximately 50 such proteins , the largest group of 27 being encoded by microexon genes . A second group comprised hydrolytic enzymes that operate at an acid pH . We showed by hybridisation experiments that expression of these genes is indeed localised to either the anterior or the posterior gland . We have suggested that this complex mixture of secreted proteins act together to perform the biological processes that occur in the lumen or , in the case of O-glycosylated membrane proteins , form a protective lining coat . We now want to discover which of them can serve as immune targets in infected animal hosts .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
The Schistosome Esophagus Is a ‘Hotspot’ for Microexon and Lysosomal Hydrolase Gene Expression: Implications for Blood Processing
|
The productive human papillomavirus ( HPV ) life cycle is tightly linked to the differentiation and cycling of keratinocytes . Deregulation of these processes and stimulation of cell proliferation by the action of viral oncoproteins and host cell factors underlies HPV-mediated carcinogenesis . Severe HPV infections characterize the wart , hypogammaglobulinemia , infection , and myelokathexis ( WHIM ) immunodeficiency syndrome , which is caused by gain-of-function mutations in the CXCR4 receptor for the CXCL12 chemokine , one of which is CXCR41013 . We investigated whether CXCR41013 interferes in the HPV18 life cycle in epithelial organotypic cultures . Expression of CXCR41013 promoted stabilization of HPV oncoproteins , thus disturbing cell cycle progression and proliferation at the expense of the ordered expression of the viral genes required for virus production . Conversely , blocking CXCR41013 function restored virus production and limited HPV-induced carcinogenesis . Thus , CXCR4 and its potential activation by genetic alterations in the course of the carcinogenic process can be considered as an important host factor for HPV carcinogenesis .
Human papillomaviruses ( HPVs ) are a family of highly related non-enveloped epitheliotropic viruses that have co-evolved with their human host and developed powerful strategies to establish persistent infection [1] . Many reports have described HPVs as commensal viruses that can persist on healthy skin [2–4] . HPVs selectively infect basal keratinocytes of stratified epithelia and other discrete populations , including the cells located in the squamocolumnar junction of the cervix [5 , 6] . These viruses undergo productive replication strictly in the terminally differentiated layers of the infected epithelium , and in most cases , cause no tissue damage or only benign warts [7] . Although host immune responses resolve most infections , instauration of persistent infections by the mucosal types of HPVs classified as high-risk , of which HPV16 and HPV18 are the most significant , is responsible for almost all cases of cervical carcinoma , a leading cause of cancer death in women . These viruses are also responsible for most anal cancers , as well as a fraction of vulval , vaginal , penile , and oropharyngeal cancers , causing nearly 5% of human cancers worldwide [8] . Cutaneous high-risk HPV types that normally persist without symptoms have also been associated with non-melanoma skin cancers in some rare genetic diseases or in immunosuppressed patients [9 , 10] . The oncogenicity of high-risk HPVs appears in the context of asymptomatic persistent infections that can hinder host immune responses as a result of evasion and subversion strategies [11] . In support of this idea , clinical observations suggest that the frequency of HPV-associated cancers is increased in immunosuppressed patients [12 , 13] . HPV-associated cancers are generally non-productive infections in which the viral E6 and E7 oncoproteins are abnormally expressed . In contrast , the timing and induction of E6 and E7 expression are tightly controlled during productive HPV infection , as the infected cells migrate towards the epithelial surface where the sequential appearance of the E4 , L2 , and L1 viral proteins allows virus release [14] . Although HPV oncoproteins are primarily responsible for the initiation and progression of cancer , only a small percentage of people infected with high-risk HPVs develop cancers . This indicates a contribution for host factors in HPV malignancy , although their mechanisms remain poorly understood [1] . One possible mechanism is that the changes in host cell signaling pathways that occur during disease progression arise from deregulated E6/E7 oncoprotein expression , which increases the risk of transformation [15] . HPV integration found in many HPV-positive carcinomas predominantly results in deregulated oncoprotein expression . Integration has also been associated with several recurrent genomic alterations and the elevated expression of host cell genes adjacent to the integration site [16] . Additionally , the oncoproteins of high-risk HPV types have been proposed to induce genomic instability through subversion of the cellular functions involved in DNA damage and repair responses [11 , 17] . In addition , host factors might confer high susceptibility for the development of HPV-associated cancers , as cervical intraepithelial neoplasias and cancers related to carcinogenic HPV infection have been linked to infection of the cervical reserve cells [18] and/or the discrete population of cuboidal cells , which are located in the cervical squamocolumnar junction and express a unique panel of genes [5 , 19] . Moreover , severe HPV-associated pathogenesis ( e . g . persistent verrucosis , dysplasia , neoplasia , some cutaneous squamous cell cancers , and a high prevalence of genital cancer ) is an underappreciated major manifestation of some primary immunodeficiencies and investigation of these clinical developments has provided clues to the host risk factors involved [20] . The wart , hypogammaglobulinemia , infection , and myelokathexis ( WHIM ) syndrome is associated with inherited gain-of-function mutations in the CXCR4 gene that encodes a receptor for the CXCL12 chemokine [21] . Binding of CXCL12 to CXCR4 triggers typical activation of Gαi protein-dependent pathways of a chemokine receptor that are regulated in a timely manner by β-arrestins , which preclude further G protein activation ( i . e . , desensitization ) and also link CXCR4 to additional signaling pathways involved in cytoskeleton reorganization and anti-apoptotic signaling [22 , 23] . In WHIM , the CXCL12/CXCR4 signaling pathways manifest by abnormally increased and prolonged G protein- and β-arrestin-dependent responses associated with an impaired desensitization of CXCR4 . Such dysfunction are responsible for the characteristic panleukopenia [24] in WHIM patients and likely also account for the severity of HPV disease through mechanisms that involve target cells and systemic immunity . From an immunological perspective , the unprecedented remission of HPV-induced warts in a WHIM patient after spontaneous partial inactivation of CXCR4 and subsequent restoration of immune function [25] suggests a role for myeloid cells in HPV life-cycle in support of the anomalies observed in WHIM patients’ myeloid cells [26] . However , whether it indicates that myeloid cells participate to host defense against HPV or , that they contribute to HPV-induced disease , when altered in WHIM patients , as reported in instances of chronic inflammation [27] , remains unknown . Evidence for the involvement of the CXCL12/CXCR4 pair in the HPV life cycle arose from the abnormal and specific expression of CXCL12 observed in keratinocytes of HPV-productive skin or mucosal lesions regardless of whether patients suffer from WHIM [28] . Expression levels of CXCL12 and its receptors , which increase in keratinocytes as a consequence of HPV genome expression , generate an autocrine signaling loop essential for keratinocyte proliferation and migration . This interplay is involved in the WHIM-associated gain-of-function CXCR4 mutant , which confers transforming capacity on HPV18-immortalized keratinocytes in mice [29] . This further supports the hypothesis that dysfunction of the CXCL12/CXCR4 signaling pathway contribute to the pathogenesis of HPV-associated cancer . However , the mechanism accounting for this process and whether it affects viral replication is not known . Here , we investigated a possible role for the WHIM-associated gain-of-function mutant receptor CXCR41013 in the HPV life cycle using HPV18 in the context of three-dimensional organotypic raft cultures of keratinocytes , the sole replicative model for HPV [30] . Our data indicate that the CXCR41013 mutant receptor shifts the HPV18 life cycle toward carcinogenesis , which is reflected in a viral gene expression pattern that favors oncoprotein stabilization . Blockade of either the CXCL12/CXCR4 axis or downstream effectors restored the productive HPV life cycle . Thus a main finding is that the WHIM-associated mutant CXCR4 receptor has a keratinocyte-intrinsic effect on HPV life cycle , supporting the idea that not all the effects of this mutation are mediated through dysregulation of the immune system . Besides , our results demonstrate an important function for the CXCL12/CXCR4 axis in the control of keratinocyte proliferation and its role in triggering transformation by HPV upon signaling imbalances resulting from cell hyperproliferation and elevated levels of oncoprotein-induced signaling .
Expression of CXCR41013 but not the wild-type CXCR4 receptor ( CXCR4wt ) confers primary human keratinocytes immortalized by HPV18 with transforming capacity , such that they develop solid tumors in nude mice [29] . We therefore explored a role for CXCR41013 in the HPV vegetative cycle in experimental models . As HPV properly replicates only within stratified epithelium , we set up three-dimensional organotypic cultures ( raft cultures ) that form dermal and epidermal layers resembling those in skin . These cultures support the productive HPV life cycle because the infected epithelial cells differentiate during their migration towards the epithelial surface . We used spontaneously immortalized human keratinocytes ( NIKS cells ) that grow normally , and can undergo terminal differentiation when cultured in rafts [31] . In NIKS cells the expression levels of ectopically expressed wt and mutant CXCR4 receptors ( CXCR4wt or CXCR41013 ) were in the same range at the RNA and protein levels ( S1 Fig , panels A , B and C ) . Viral genome copy numbers were similar among the various NIKS cells in monolayer ( i . e . expressing the endogenous CXCR4wt solely or with either CXCR4wt or CXCR41013 ectopically expressed receptors ) and were increased in the same range upon differentiation of NIKS cells in organotypic cultures ( S2 Fig , panel A ) . Viral transcript levels ( E6E7 and E2 ) were also similar in NIKS cells expressing the exogenous wt or mutant forms of the receptor ( S2 Fig , panel B ) . The global architectures and subcellular topologies of raft cultures expressing CXCR41013 or CXCR4wt ( CXCR41013- or CXCR4wt-rafts ) was apparently unchanged within CXCR41013-expressing raft cultures ( Fig 1 , panel A ) as well as the expression pattern of CXCR4 receptors ( endogenous CXCR4 and ectopically expressed wt and mutant forms ) in basal and supra-basal layers ( S1 Fig , panel D ) . CXCR41013- and CXCR4wt-rafts also exhibited normal expression patterns for keratin 10 in the spinous and granular layers , and the keratin filament-associated filaggrin protein in the granular layer , which are two prominent markers of epidermal differentiation ( Fig 1 , panels B-C ) . In contrast , replication of the viral genome was strongly diminished in CXCR41013-raft cultures compared to that in CXCR4wt-rafts ( Fig 2 , panel A ) . Consistent with its role in HPV replication [32] , production of the E2 viral protein was also lower in CXCR41013-raft cultures ( Fig 2 , panel B ) . Production of viral proteins involved in the completion of the HPV life cycle in the upper epithelial layers , were also dramatically reduced ( L1 ) or nearly undetectable ( E4 ) in CXCR41013-rafts compared to their levels in CXCR4wt-rafts ( Fig 2 , panels C-D ) . The E4 expression pattern in native NIKS raft cultures expressing endogenous CXCR4 ( S3 Fig ) was comparable to that observed in CXCR4wt-raft cultures ( Fig 2 , panel C ) , further supporting the relevance of the CXCR4wt-raft model . We have then searched for the biosynthesis of infectious viral particles , which is the final step in the HPV life cycle . We found that CXCR4wt-rafts are producing virions , which were infectious in an HaCaT cell infection assay as detected by the presence of the HPV spliced E1E4 transcript [33] , while the CXCR41013-rafts did not contain detectable infectious virions ( S4 Fig ) . These data were correlated with the presence of koilocytic cells in the intermediate layers of the CXCR4wt-rafts ( e . g . Fig 1 , panels A-B ) . Collectively , the dramatic lowest production of E2 , L1 , and E4 proteins in CXCR41013-rafts together with the absence of any detectable infectious virions strongly suggest that this cell environment restricts replication and the productive HPV life cycle . To gain further insight into the consequences of CXCR41013 expression on HPV life cycle , we analyzed expression of the E6 and E7 viral oncogenes and their surrogate markers . Western blot analyses of raft cultures showed that E6 and E7 protein production in CXCR41013-rafts was significantly higher than in CXCR4wt-rafts ( Fig 3 , panel A and control experiments for antibodies specificity in S5 Fig ) . Since we were unable to detect E7 and E6 proteins in raft cultures by immunohistochemistry because of high background , we used surrogate markers to investigate the expression of these proteins . E7 is known to disrupt proteins involved in cell cycle progression , resulting in substantial induction of a functionally inactive form of cyclin-dependent kinase inhibitor 2A ( p16 ) , which can be used as an indicator of deregulated E7-expression and HPV-associated dysplastic and neoplastic lesions [9] . The minichromosome maintenance ( MCM ) family of cell cycle proteins are essential for eukaryotic DNA replication , and MCM7 and MCM2 are widely used as molecular surrogates of E6/E7 expression and E6/E7-mediated cell cycle entry [34] . In agreement with this , MCM2 and p16 expression levels were higher in CXCR41013-rafts than in CXCR4wt-rafts . Compared to the limited distribution of p16 staining in basal layers , MCM2 staining was more widespread and intense , and extended into the upper epithelial layers ( Fig 3 , panels B-C-D ) . The significantly higher levels of oncogene expression observed in CXCR41013-rafts , were not paralleled by neither higher E6/E7 transcript levels nor changes in E2 transcripts levels ( S6 Fig , panel A ) . This was consistent with our inability to detect any HPV genome integration ( S6 Fig , panel B ) or enhanced transcriptional activity of the HPV long control region ( LCR ) ( S6 Fig , panel C ) in CXCR41013-rafts . To determine whether the E6 and E7 oncoproteins were stabilized in CXCR41013 cells , we quantified E6 and E7 protein levels over time in CXCR4wt- and CXCR41013-expressing keratinocytes induced to differentiate in medium containing a high concentration of calcium . This model permits the activation of late events and the productive phase of the HPV life cycle after 48−96 h of culture [35] . In NIKS cells differentiated for 96 h in high-calcium medium , E6 and E7 protein decay after 2 h of cycloheximide treatment was lower in CXCR41013 keratinocytes than in CXCR4wt keratinocytes , resulting in higher relative levels of E6 and E7 in CXCR41013 keratinocytes ( S7 Fig , panel A ) . In contrast , in undifferentiated NIKS cells we observed no significant difference in the stability of oncogenes ( S7 Fig , panel B ) . These results suggest that the viral oncoproteins were stabilized in the presence of CXCR41013 in differentiated cells . An elevated proliferation rate is an early step in viral oncogenesis [8] . The Ki-67 antigen , which is expressed in all phases of the cell cycle except in G0 , is widely used to assess proliferation and represents a biomarker for cervical cancer [34] . Immunohistochemical analyses of Ki-67 in raft cultures indicated higher levels of cellular proliferation in the basal and suprabasal compartments of CXCR41013 rafts than in those of CXCR4wt rafts , and a higher overall proportion of ( Ki-67-positive ) cells in cycle ( Fig 4 panels A , C ) . Keratinocyte proliferation is normally restricted to the basal layers in CXCR41013-rafts in the absence of HPV infection ( S8 Fig ) , indicating that CXCR41013 does not deregulate cell proliferation on its own but rather acts synergistically with HPV . Given the capacity of E6 to promote degradation of the p53 protein , a gatekeeper of aberrant cell cycle progression involved in cell cycle arrest and apoptosis [36] , we quantified apoptotic cell numbers in raft sections by TUNEL assay and p53 protein levels by western blot ( Fig 4 ) . There were fewer TUNEL-positive apoptotic cells in sections of CXCR41013 rafts than in CXCR4wt rafts suggesting that the overall proportion of apoptotic cells can significantly lower in CXCR41013 rafts ( Fig 4 , panels B-C ) . Consistent with these results , p53 levels were significantly lower in CXCR41013-rafts than in CXCR4wt rafts ( Fig 4 , panel D ) . These results demonstrate that CXCR41013 plays an important role in driving the HPV18 viral life cycle toward carcinogenesis , notably by increasing the levels of the HPV oncoproteins . HPV , like other viruses associated with human cancers , was recently shown to hijack DNA damage response pathways responsible for maintaining genomic integrity . Proteins involved in these pathways include the ataxia telangiectasia mutated ( ATM ) and Rad3-related protein kinases , as well as p53 and the CHK1 and CHK2 kinases , which are involved in downstream checkpoint pathways [17] . Activation of the ATM/CHK pathway in the course of HPV infection provides a suitable environment for viral replication in differentiated cells [37] . The mechanisms of this activation , which may be part of the HPV replication process itself , are not completely understood but may involve the E1 , E2 , and E7 proteins [38] . To investigate whether the ATM/CHK pathway can be differentially modified in a CXCR41013-expressing background as a result of altered viral replication or higher E7 expression levels , we analyzed the expression of the ATM and CHK proteins in keratinocytes differentiated in high-calcium medium . Progression of normal differentiation was indicated by increases in involucrin protein abundance over time in CXCR41013- and CXCR4wt-expressing NIKS cells cultured in high-calcium medium ( Fig 5 , panel A ) . In differentiated HPV18-positive NIKS cells , E6 protein levels were higher in cells expressing CXCR41013 than in those expressing CXCR4wt ( Fig 5 , panel B ) . These results confirmed and extended our findings in raft cultures ( Fig 3 ) . Confirming our results in raft cultures ( Fig 4 ) , p53 protein levels at 96 h were lower in CXCR41013-expressing cells than in CXCR4wt-expressing cells ( Fig 5 , panel A ) , which was likely related to the higher levels of E6 protein in CXCR41013-expressing cells . The expression patterns of ATM and its activated phosphorylated form ( pATM ) in control cultures ( Fig 5 , panel C , CXCR4wt-cells ) were concordant with previous report [37] . After 96 h of culture , pATM levels were lower in CXCR41013-expressing cells than in CXCR4wt-expressing cells ( Fig 5 , panel C ) . Accordingly , the levels of the activated phosphorylated form of CHK2 ( pCHK2 ) kinase were significantly lower in keratinocytes expressing CXCR41013 than in those expressing CXCR4wt at 96 h but also at 48 h ( Fig 5 , panel C ) . This correlates with the tendency of CXCR41013-expressing cells to have lower levels of activated ATM . Altogether these results suggest that the presence of CXCR41013 might lead to suboptimal activation of the ATM pathway and a reduced ability to support HPV replication . The suboptimal HPV replication environment in keratinocytes expressing CXCR41013 prompted us to examine the consequences of normalizing the CXCR41013-enhanced signaling . One of the key pathways accounting for the CXCR41013 gain-of-function is the enhanced β-arrestin-mediated signaling , which is dependent upon the third intracellular loop ( SHSK motif ) of CXCR4 [39] , which is also required for mobilization of intracellular calcium and G-protein-independent stimulation of JAK2/STAT3 in response to CXCL12 [40] . Therefore , we expressed CXCR41013 lacking the SHSK motif , CXCR41013&ΔSHSK in keratinocyte raft cultures ( Fig 6 ) . In leukocytes , CXCR41013&ΔSHSK exhibited normal β-arrestin–mediated signaling and CXCL12-induced chemotaxis [39] . In raft cultures expressing CXCR41013&ΔSHSK , E2 protein levels were significantly higher than in CXCR41013-rafts ( Fig 6 , panel A ) . E4 and L1 proteins were detected in the upper epithelial layers of the CXCR41013&ΔSHSK-rafts ( Fig 6 , panels B-C ) . Proliferating cells expressing the Ki-67 antigen remained confined to basal and suprabasal layers ( Fig 6 , panel D ) , as in CXCR4wt raft ( Fig 4 , panel A ) . Thus , CXCR41013&ΔSHSK , with normal β-arrestin-mediated signaling , lacks the ability to impair virus production , further supporting the capacity of CXCR4 to tune the viral life cycle through its downstream signaling pathways . To further assess the role of CXCR4 activity in the HPV life cycle , CXCR4wt- and CXCR41013-rafts were treated with AMD3100 , a selective and competitive antagonist of CXCR4 [41] that efficiently blocks CXCR41013 function [28] . The architecture and viral expression patterns of raft cultures treated with AMD3100 were compared to control ones ( untreated rafts ) by immunohistochemistry and western blot analyses ( Fig 7 ) . The stratified epithelium in CXCR4wt and CXCR41013-rafts treated with AMD3100 was thinner than in untreated rafts ( Fig 7 , panel A ) but keratinocytes differentiation was apparently not affected given the normal expression pattern for keratin 10 ( S9 Fig panel A ) . E4 and L1 proteins were readily detected in the upper layers of CXCR41013-rafts treated with AMD3100 ( Fig 7 , panel B ) as in CXCR4wt-rafts treated with AMD3100 ( S9 Fig panel B ) , while sparsely detected ( L1 ) or undetectable ( E4 ) in untreated CXCR41013-raft cultures ( Fig 7 , panel B ) . In contrast , AMD3100 treatment reduced expression of the E6 and E7 oncoproteins in CXCR41013- and CXCR4wt-rafts ( Fig 7 , panel C and D , respectively ) . Thus blocking CXCR41013-dependent signaling allows virus production , while reducing oncogene-expression and–driven neoplastic-like changes . Moreover , we found that tumors produced by injecting CXCR41013-expressing human keratinocytes immortalized by HPV18 [29] into nude mice were significantly smaller in AMD3100-treated mice than in untreated mice ( Fig 8 ) . Collectively , these results indicate that CXCL12/CXCR4 signaling impacts the balance between HPV replication and HPV-driven carcinogenesis .
Considering the role of the CXCL12/CXCR4-signaling in the migration and survival of HPV18-infected keratinocytes , we investigated the importance of this axis in the productive HPV life cycle in which the timely and coordinated expression of different viral genes occurs as infected cells move toward the epithelial surface where virions mature . In view of the consequences of axis deregulation in HPV-induced cell transformation via the CXCR41013 gain-of-function mutant , we investigated the impact of CXCR41013 expression on the productive HPV life cycle in an organotypic model of human epidermis . Our results demonstrate that unregulated CXCR41013 function fosters a keratinocyte environment that restrains the productive HPV18 life cycle in contrast to CXCR4wt-expressing rafts that efficiently reproduce the complete HPV life cycle including the production of infectious virions . These findings in CXCR41013-rafts correlate with deregulated keratinocyte proliferation and the stabilization of the E6 and E7 oncoproteins together with dramatic decreases in production of the late viral proteins involved in HPV virion assembly . These results are especially relevant because CXCL12 is expressed in keratinocytes from HPV-infected raft cultures ( S10 Fig ) . This extends previous studies detecting CXCL12 expression in epidermal keratinocytes from HPV-induced lesions but not from other skin pathologies and normal skin [28 , 42] . Thus collectively , our results support the concept that the CXCL12/CXCR4 pathway controls the HPV productive life cycle , likely reflecting the normal function of this pathway as a regulator of keratinocyte proliferation and survival . When deregulated , as in the WHIM syndrome , the CXCL12/CXCR4 pathway can trigger HPV-induced transformation as a result of elevated levels of oncoprotein-induced signaling and down-regulation of DNA damage response pathways associated with cell hyperproliferation . On the one hand , the severe pathogenesis in WHIM patients manifests as intractable genital warts that often develop into severe dysplasia and carcinoma [43 , 44] . This pathogenesis is generally due to mucosal high-risk HPV types for which we can suggest that expression of CXCR41013 might enhance transforming capacity , partly through the elevation of E6/E7 proteins expression that drive proliferation of the infected cells , although this remains to be investigated directly in patients-derived cells . In some cases , patients’ dysplasia was found to be associated with low-risk HPV types , such as HPV6 , which display potential oncogenicity [45] . On the other hand , the profusion and persistency of cutaneous warts is another major clinical manifestation of the WHIM-associated HPV pathogenesis . In this regard , it can be postulated a role for CXCR41013 in driving the initial proliferation of the infected cells , thus allowing expression and replication of cutaneous low-risk HPV types , which do not normally stimulate proliferation . Our study is providing the rational for future mechanistic investigations of the productive life cycle of cutaneous low-risk HPV , which remains underappreciated due to the lack of robust in vitro models . We have found that CXCR41013 expression in keratinocyte raft cultures is associated with the stabilization of the E6 and E7 oncoproteins . This was not associated with an integration of the HPV genome further supporting that disruption of the E2 ORF is not the only mechanism of suppressing E2 and increasing E6 and E7 expression as previously reported in HPV16-induced carcinogenic progression [46] . Such modulations of oncoproteins expression might involve the ubiquitin-proteasome system , which is involved in the degradation of E6 produced by the high-risk types HPV18 and HPV16 [47] . Additionally , the capacity of E6 to interact with certain PDZ domain proteins and phosphoserine-binding proteins involved in cell signaling pathways is a mechanism for regulating E6 stability that is common to diverse high-risk HPV types [48–50] . Beside PDZ domain proteins , chemokine receptors including CXCR4 can interact with chaperone proteins , some being recently found to increase the steady-state levels and half-life of E6 and E7 oncogenes [51 , 52] . As the interaction of E6 with either of these protein families depends on its phospho-regulation by various kinases ( e . g . protein kinase A or B ) and determines the fate of E6 , different environmental conditions might have a significant impact on the likelihood of HPV infection progressing toward malignancy [53] . We propose that changes in cell signaling pathways in the context of CXCR41013 expression , and notably in the downstream kinases activated by CXCL12-CXCR4 signaling may differentially affect the stability of HPV18 E6 . Some kinases , as well as the rate of ubiquitination , can also control the steady-state level of E7 by interfering with proteasome-dependent degradation , but these processes have been studied in the context of only a few HPV types ( HPV16 and HPV6 ) [54 , 55] . Increased production of E6 and E7 proteins in the context of CXCR41013 expression makes rafts prone to drive the viral lifecycle toward carcinogenesis as demonstrated by the altered levels of keratinocyte proliferation and apoptosis and by a disturbance in the ordered expression of viral gene products that normally leads to virus replication and production . Whether the fact that rafts derive from the spontaneously immortalized human NIKS keratinocytes might contribute to this process is not known and is awaiting the setting of rafts from primary human keratinocytes for HPV life cycle modeling . Previous studies have clearly suggested that an increase in high-risk HPV protein levels can drive a more severe neoplastic phenotype [56] . Among these viral proteins , we observed a dramatic decrease in the level of E2 that might be related to enhanced degradation by the ubiquitin-proteasome system , which controls viral protein stability and was proposed to operate in cycling cells [57] . Aberrant cell cycle progression in CXCR41013-rafts might thus increase E2 degradation and diminish its activity in the late phases of the viral life cycle . Conversely , normalizing arrestin-dependent signaling downstream of CXCR41013 ( CXCR41013&ΔSHSK ) might stabilize E2 protein thus partially restoring the productive HPV life cycle . This shift toward viral production was revealed by the enhanced production of L1 and E4 , which are primarily involved in genome packaging and virus release [1 , 58 , 59] . The stabilization of E2 in CXCR41013&ΔSHSK-rafts might also be accounted for the physical and functional interaction of E2 and E4 as the level of each protein is increased by the presence of the other [60] . Additionally , and consistent with the recently reported essential role of DNA damage responses in the viral replication [37] , ATM/CHK pathway activation was found to be significantly lower in CXCR41013-rafts . The contribution of ATM/CHK pathway activation to viral replication has begun to be deciphered but its function in HPV-induced cancer development remains unclear , especially because E7 and E6 have important roles in promoting this activation [38 , 61] . Decreased activation of the ATM/CHK proteins in CXCR41013-rafts , in spite of increased expression of E6 and E7 , may appear paradoxical . However , such deregulation might include the JAK/STAT pathway , which is induced downstream of CXCR4 in a G protein-independent manner [62] and was shown to activate the ATM-dependent DNA damage responses [38] . Although the biological role of DNA damage responses in HPV-induced malignancy is still uncertain , decreasing the production or activation of ATM/CHK proteins would likely lead to the accumulation of DNA damage in CXCR41013-rafts . In cervical disease , it is thought that the levels of E6 and E7 rise with cervical intraepithelial neoplasia ( CIN ) severity . Changes in gene expression underlie the neoplastic progression , with CIN1 lesions supporting the complete HPV life cycle in contrast to CIN3 lesions that are considered to be high-grade precancerous . We provide evidence that blocking CXCL12/CXCR4-dependent signaling in the course of raft culture differentiation allowed the productive HPV life cycle to proceed at the expense of the HPV-induced carcinogenesis in CXCR41013-rafts . These results provide molecular clues to the potential therapeutic effect of AMD3100 treatment for skin warts when combined with imiquimod [63] but also for HPV-induced carcinogenesis in a mouse preclinical model [64] and here in nude mice after injection of human keratinocytes immortalized by HPV18 and expressing CXCR41013 . Whereas we think that the interplay between the HPV life cycle and CXCL12/CXCR4 in keratinocytes is the direct mediator element , the beneficial effect of AMD3100 might also be related to other components that are controlled by this signaling axis ( e . g . resident and infiltrating immune cells in the skin , stem cell recruitment , or endothelial cell responses ) . The CXCL12/CXCR4 pair is indeed involved in the increased survival and/or proliferation of cancer cells from various types including virus-related cancers , as well as in the promotion of tumor metastasis and angiogenesis linked to tumor progression [65–69] . In light of the HPV-pathogenesis associated with the WHIM syndrome , it can be extrapolated that dysfunction of CXCR4 might be acquired from genetic errors accumulated during the multistep process of HPV-induced neoplasia , as demonstrated by the somatic WHIM-like CXCR4 mutations reported in Waldenström macroglobulinemia [70] or the anomalies in the effectors of the CXCL12/CXCR4-signaling pathway reported in patients with GATA2-deficiency [71–73] . Clues to the additional effectors of this potentially pathogenic process arise from the abnormal expression of CXCL12 in HPV-lesions in the general population of HPV infected individuals [28] and from the presence of CXCL12 ( our data ) among the panel of expressed genes unique to the discrete population of cuboidal cells located in the cervical squamocolumnar junction , which have been linked to HPV-related cervical intraepithelial neoplasia and cancers [5] . Consequently , the CXCL12/CXCR4 signaling pathway appears to be an important host factor in HPV-induced pathogenesis .
NIKS cells , near-diploid spontaneously immortalized human keratinocytes [31] ( kindly provided by Dr Paul F . Lambert ) , were maintained at subconfluence on mitomycin C-treated 3T3-J2 feeder cells ( kindly provided by Dr Paul F . Lambert ) in F medium with all supplements as previously described [30] . Human foreskin fibroblasts ( kindly provided by Dr Paul F . Lambert ) were cultured in Ham’s F12 medium containing 10% fetal bovine serum and 1% penicillin-streptomycin , before use in raft cultures . NIKS cells expressing CXCR4wt , CXCR41013 or CXCR41013&ΔSHSK were obtained with a lentivirus-mediated strategy as previously described [29 , 39] . Expression of similar levels of each receptor in the different cell populations was checked by flow cytometry ( S1 Fig , panel B ) . Recircularized HPV18 DNA was prepared as previously described [30] . The different NIKS cell populations ( 2 . 5 x 106 cells plated the day before the transfection ) were cotransfected with 2 μg of recircularized HPV18 DNA and 0 . 5 μg of the blasticidin-resistance plasmid pcDNA6 using Effectene Transfection Reagent ( QIAGEN ) . Blasticidin selection ( 7 μg/mL ) was performed for 6 days . HPV18-positive NIKS cells expressing CXCR4wt , CXCR41013 , or CXCR41013&ΔSHSK were grown in raft cultures to induce the three-dimensional architecture of the stratified epithelium , as described previously [30] . Briefly , 1 . 5 x 106 NIKS cells were seeded onto a dermal equivalent composed of rat-tail collagen type 1 containing 1 x 106 human foreskin fibroblasts . Raft cultures were lifted onto transwell inserts submerged in deep well plates containing keratinocyte plating medium and cultured for 4 days . The transwell inserts were then raised by placing four cotton pads underneath them , thereby exposing the epithelial cells to the air-liquid interface . Raft cultures were fed by diffusion from below with cornification medium and were allowed to stratify for 14 days . When specified , AMD3100 ( A5602 , Sigma-Aldrich ) was added to the cornification medium at a concentration of 20 μg/mL from day 4 to the end of the experiment . Rafts were then removed from transwell inserts and either fixed in formalin and embedded in paraffin for histological analyses , or frozen at −80°C for quantitative real-time PCR and western blot analyses . To induce differentiation in medium containing 1 . 5 mM CaCl2 ( high calcium ) , HPV18-positive NIKS cells expressing either CXCR4wt or CXCR41013 were cultured in absence of 3T3-J2 feeders in F medium for 24 h and then switched to F medium ( without growth supplements ) containing 1 . 5 mM CaCl2 . Cells were then harvested at 0 , 48 , and 96 hours for protein extraction . NIKS cells from monolayer cultures or rafts cultures were resuspended in protein lysis buffer ( 1% Triton X-100 , 10 mM Tris-HCl pH 7 . 4 , supplemented with Protease and Phosphatase Inhibitor Tablets ( Pierce ) ) . Protein concentration was measured with the BCA Protein Assay Kit ( Pierce ) according to the manufacturer’s protocol . Equivalent amounts of protein were separated on a SDS-polyacrylamide gel and transferred to a PVDF membrane . Primary antibodies were as follows: anti-HPV18-E6 ( kindly provided by Arbor Vita Corporation ) , anti-HPV18-E7 ( sc-1590 , Santa Cruz ) , anti-HPV18-E2 ( kindly provided by Dr . F . Thierry ) , anti-p53 ( sc-126 , Santa Cruz ) , anti-involucrin ( I9018 , Sigma-Aldrich ) , anti-GAPDH ( 14–9523 , eBioscience ) ; anti-pATM ( Ser1981; #13050 ) , anti-ATM ( #2873 ) , and anti-pCHK2 ( Thr68 ) and anti-CHK2 ( #2661 and #3440 , respectively , Cell Signaling ) . Membranes were incubated with the appropriate secondary antibodies conjugated to HRP ( GE Healthcare ) . Proteins were detected using the Immobilon Western Chemiluminescent HRP kit ( Millipore ) . Raft paraffin sections ( 5-μm thick ) were stained with hematoxylin and eosin ( HE ) , and with safran ( HES ) where indicated . Immunohistochemistry was performed on paraffin sections using primary antibodies for keratin 10 ( MA1-35540 , Thermo Scientific ) , filaggrin ( VP-F706 , Vector laboratories ) , HPV18-E2 ( provided by Dr . F . Thierry ) , HPV18-E4 ( provided by Dr . J . Doorbar ) , HPV18-L1 , p16 ( sc-56330 , Santa Cruz ) , Ki-67 ( 18-0191Z , AbCys ) , and CXCL12 ( K15C clone , MABC184 , EMD Millipore ) . Bound antibodies were detected using the LSAB+/HRP kit ( K0679 , Dako ) or the AEC+ High Sensitivity Substrate Chromogen Ready-to-Use System ( K3461 , Dako ) . For immunofluorescence , staining with the primary antibody for MCM2 ( ab31159 , Abcam ) was followed by staining with a goat anti-rabbit Alexa Fluor 596 ( Invitrogen ) . Tissues were counterstained with DAPI . For the TUNEL assay to detect apoptotic cells , we used the In Situ Cell Death Detection Kit , Fluorescein ( 11684795910 , Roche Diagnostics ) according to the manufacturer’s instructions . In situ hybridization was performed with the wide spectrum HPV biotinylated DNA probe sets able to detect 11 types of anogenital HPV ( In Situ Hybridization Detection System , K0601 , Dako ) . Where indicated , slides were scanned by the digital slide scanner NanoZoomer 2 . 0-RS ( Hamamatsu ) allowing an overall view of the samples . Images were digitally captured from the scanned slides using the NDP . view2 software ( Hamamatsu ) . All quantifications from the histological analyses were performed by counting 10 different fields on the scanned slides . Other slides were analyzed using a Leica DMLA microscope , in particular to visualize images at 100X magnification , and images were captured with a Leica DFC450 C digital microscope camera . Immunochemical stainings were interpreted simultaneously and independently by at least two investigators ( FM , LC , or FG ) . HK-HPV18-CXCR41013 tumors were established in nude mice as previously described [29] . Briefly , athymic female nude nu/nu 5-week-old mice ( Harlan Laboratories ) were injected subcutaneously with 2 x 107 HK-HPV18 cells expressing T7-GFP-tagged CXCR41013 in the right flank ( six to seven mice per group ) . Mice were treated with 5 mg/kg of AMD3100 ( intraperitoneal administration ) on days 8 , 13 , 16 , and 20 after tumor cell injection . Tumor volumes ( V ) were calculated as V = π/6 x ( length x width2 ) . All experimental procedures were conducted in our animal facility ( agreement n° B 92-023-01 ) in accordance with the European Union’s legislation and the relevant national legislation , namely the French “Décret no 2013–118 , 1er février 2013 , Ministère de l’Agriculture , de l’Agroalimentaire et de la Forêt” regarding the use of laboratory animals and were approved by the Committee on the Ethics of Animal Experiments ( Comité d'Ethique en Expérimentation Animale Capsud or CEEA-26 ) under the authorization 2014_039 #2521 . Student’s t test was used to compare the significance between specified groups . All analyses were performed with GraphPad Prism software .
|
Human papillomaviruses ( HPV ) are epitheliotropic tumor viruses causing mostly benign warts but that have developed strategies to establish persistent infections . Although host immune responses clear most infections , persistence of some HPV types causes ~5% of human cancers and severe pathogenesis in immunosuppressed individuals . How early events in HPV infection , determined by the interaction between viral and host proteins , might lead to viral persistence and pathogenesis is unknown . Here , we thought to investigate this issue by providing mechanistic insights into the selective susceptibility to HPV pathogenesis displayed by patients who are immunosuppressed as a consequence of mutations in the CXCR4 gene encoding for the receptor of the CXCL12 chemokine ( WHIM syndrome ) . We previously unraveled the existence of a general interplay between the CXCL12/CXCR4 axis and HPV , which is hijacked toward cell transformation upon expression of the CXCR4 mutant . Here , using three dimensional epithelial cell cultures to analyze the HPV life cycle , we found that the CXCR4 mutant promotes cell hyperproliferation and stabilization of viral oncoprotein expression at the expense of virus production . Our results , which identify CXCR4 as an important gatekeeper of keratinocyte proliferation and as a new susceptibility factor in HPV pathogenesis , may be translated into anti-viral and anti-cancer strategies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"urology",
"cell",
"death",
"keratinocytes",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"cell",
"processes",
"microbiology",
"cell",
"differentiation",
"epithelial",
"cells",
"oncology",
"developmental",
"biology",
"sexually",
"transmitted",
"diseases",
"infectious",
"diseases",
"human",
"papillomavirus",
"infection",
"animal",
"cells",
"life",
"cycles",
"biological",
"tissue",
"pathogenesis",
"viral",
"replication",
"carcinogenesis",
"cell",
"biology",
"anatomy",
"virology",
"genitourinary",
"infections",
"apoptosis",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases"
] |
2016
|
The CXCL12/CXCR4 Signaling Pathway: A New Susceptibility Factor in Human Papillomavirus Pathogenesis
|
Antibodies that neutralize ( nAbs ) genetically diverse HIV-1 strains have been recovered from a subset of HIV-1 infected subjects during chronic infection . Exact mechanisms that expand the otherwise narrow neutralization capacity observed during early infection are , however , currently undefined . Here we characterized the earliest nAb responses in a subtype A HIV-1 infected Rwandan seroconverter who later developed moderate cross-clade nAb breadth , using ( i ) envelope ( Env ) glycoproteins from the transmitted/founder virus and twenty longitudinal nAb escape variants , ( ii ) longitudinal autologous plasma , and ( iii ) autologous monoclonal antibodies ( mAbs ) . Initially , nAbs targeted a single region of gp120 , which flanked the V3 domain and involved the alpha2 helix . A single amino acid change at one of three positions in this region conferred early escape . One immunoglobulin heavy chain and two light chains recovered from autologous B cells comprised two mAbs , 19 . 3H-L1 and 19 . 3H-L3 , which neutralized the founder Env along with one or three of the early escape variants carrying these mutations , respectively . Neither mAb neutralized later nAb escape or heterologous Envs . Crystal structures of the antigen-binding fragments ( Fabs ) revealed flat epitope contact surfaces , where minimal light chain mutation in 19 . 3H-L3 allowed for additional antigenic interactions . Resistance to mAb neutralization arose in later Envs through alteration of two glycans spatially adjacent to the initial escape signatures . The cross-neutralizing nAbs that ultimately developed failed to target any of the defined V3-proximal changes generated during the first year of infection in this subject . Our data demonstrate that this subject's first recognized nAb epitope elicited strain-specific mAbs , which incrementally acquired autologous breadth , and directed later B cell responses to target distinct portions of Env . This immune re-focusing could have triggered the evolution of cross-clade antibodies and suggests that exposure to a specific sequence of immune escape variants might promote broad humoral responses during HIV-1 infection .
Protective vaccines against viral infections generally elicit nAb responses that are comparable to those in natural infections [1] . It is , therefore , widely accepted that an optimal vaccine against HIV-1 will need to produce nAbs , but features such as the high genetic diversity and mutability of HIV-1 Env pose unique obstacles . While broad neutralization of HIV-1 will likely be difficult to achieve through immunization , renewed optimism exists because of breakthroughs in the HIV-1 vaccine and nAb research fields . In the recently concluded RV144 vaccine trial , modest protection from acquisition of infection was observed and correlated with high levels of antibodies that recognized the V1V2 hypervariable domain of Env gp120 [2] . To date , these anti-V1V2 antibodies are the only immune correlate of vaccine-mediated protection against HIV-1 in humans . In non-human primate models , Barouch et al . reported that strong vaccine-induced protection against a diverse simian immunodeficiency virus ( SIV ) challenge in rhesus macaques correlated with V2-binding antibody titer along with nAb titers against two neutralization-sensitive heterologous SIV Envs [3] . Taken together , these results support the concept that antibodies are important for protection against HIV-1 infection and lead to the hypothesis that even higher vaccine efficacy could be achieved if broad nAbs can be induced [4] . The latest intensified efforts to recover and characterize potent and broad mAbs from chronically infected subjects with exceptional neutralization breadth have yielded important clues regarding how these mAbs overcome Env diversity . Such cross-clade neutralizing mAbs have been shown to target conserved elements in the CD4 binding site ( CD4bs ) ( e . g . VRC01 , PGV04 ) , V1V2-dependent and trimer-enhanced ( quaternary ) epitopes ( e . g . PG9 , PG16 ) , the gp41 membrane proximal external region ( MPER ) ( e . g . 4E10 , CAP206 , 10E8 ) , and glycan/V3-dependent epitopes ( e . g . PGT128 ) [5]–[11] . For each class of ‘super’ mAb , characterization of the variable domains of the immunoglobulin heavy and light chains ( VH and VL , respectively ) , in terms of their structure , germline gene utilization , level of somatic hypermutation , and the features of their heavy chain third complementarity-determining regions ( CDR H3s ) , has unveiled specific characteristics that facilitate extraordinary neutralizing capacity [8] , [12]–[15] . Importantly , substantial nAb breadth usually requires two to three years of infection to develop and occurs in only about 20–30% of infected subjects [16] , [17] . Furthermore , individuals with ‘elite’ neutralizing activity constitute only about 1% of chronically infected subjects [18] . The reasons why nAb breadth does not develop earlier or more frequently are not known but could include autoreactivity leading to clonal deletion of B cells [19] , impaired affinity maturation [20] , or induction of a particular Ig germline family [10] , [13] , [15] , [21] , [22] . It is also possible that early viral escape contributes to the process of increasing nAb breadth [23] , [24] . A paradox of neutralization breadth is that targets known to mediate this phenomenon , such as HXB2 residue N160 in V2 ( targeted by PG9 , PG16 ) or N332 near V3 ( targeted by PGT128 ) , are well conserved and present in many transmitted/founder Envs , but broad cross-clade activity only develops in a subset of individuals . The mere presence of these targets is not , then , sufficient to elicit broadly neutralizing antibodies in early infection . Here we describe the initial nAbs in a subtype A HIV-1 infection that target the N332-proximal region of gp120 that has been previously associated with broad neutralization by mAbs recovered from a chronic subtype CRF02_AG infection [6] , [8] and strain-specific nAb responses in early subtype B infection [25] . Early escape involved a single amino acid substitution in this region , which appeared to drive a modest increase in the autologous neutralization breadth of somatically related mAbs . Later escape entailed the addition and/or shifting of glycans recognized by several previously described broadly neutralizing mAbs , but these changes were not targeted by the cross-neutralizing nAbs that developed later in this subject . The combinatorial interplay among early nAb targets , viral escape pathways , and antibody somatic hypermutation could , therefore , shape the ultimate development of heterologous nAb breadth across subjects .
To examine the course and magnitude of autologous HIV-specific humoral activity in a Rwandan seroconverter , R880F , establishment and evolution of the earliest detectable nAb responses were evaluated . This subject was identified as antigen positive , antibody negative on 5Jan07 and then as antibody positive on 12Jan07 . This latter date of seroconversion was designated as the 0-month time point for our analyses . Subsequent samples were chronologically coded from this originating time point forward , and each Env was given an arbitrary letter ( A , B , or C ) and number ( 1–61 ) designation that was preceded by the time point of isolation in months post-seroconversion . For example , Envs 0-A6 and 0-B24 were singly isolated from 0-month plasma , 2-A9 and 2-A13 from 2-month plasma , 5-A5 and 5-B52 from 5-month plasma , etc . ( Table 1 ) . These viral Envs were single-genome amplified and cloned into expression plasmids for the evaluation of Env pseudotypes . The two 0-month Envs , 0-A6 and 0-B24 , had identical sequences and represented the transmitted/founder virus ( Figure S1 ) . In sum , ten envelopes from plasma at 2-months post-seroconversion , three from 5-months , five from 7-months , and two from 10-months were evaluated ( Table 1 ) . Each Env-pseudotyped virus was assayed against autologous plasma contemporaneous to its date of isolation in the Tzm-bl neutralization assay . Plasma samples from 2- , 5- , 7- , and 10-months , but not 0-months , potently neutralized the founder Envs 0-A6 and 0-B24 ( Figure 1 ) . All longitudinal Envs were at least one log less sensitive to neutralization by contemporaneous plasma than the founder Envs and were , therefore , considered humoral escape variants ( Figure 1B–E ) . These 0- , 2- , 5- , 7- , and 10-month Envs all succumbed to neutralization by plasma collected at 16-months post-seroconversion ( Figure 1F ) . Hence , the induction of de novo nAbs against contemporaneous escape variants , which we and others have previously described [26]–[30] , also occurred during the first year and a half of infection in R880F . In this subtype A HIV-1 infected subject , a potent nAb response was detected by 2-months following the first antibody positive time point and initiated repeated rounds of neutralization and viral escape . To localize the earliest nAb target and elucidate consequent mechanisms of viral escape , full-length amino acid Env sequences for all 2-month nAb escape variants shown in Figure 1B were aligned and inspected for the presence of mutational hot spots . Amino acid changes concentrated in three regions of gp120 at 2-months: in C2 immediately preceding the beginning of V3 , in the alpha2 helix in C3 , and in V5 . Figure 2 specifically diagrams these segments of gp120; Figure S1 includes the full gp160 alignment of all 22 R880F Envs . The isoleucine at position 295 ( I295; HXB2 residue 293 ) in C2 mutated to arginine ( I295R ) in two Envs or threonine ( I295T ) in one Env ( Figure 2 ) . Additionally , glutamic acid E338 in the alpha2 helix ( HXB2 residue 337 ) became three different residues including aspartic acid ( E338D ) , glycine ( E338G ) , and lysine ( E338K ) in six Envs ( Figure 2 ) . Of note , compared to the founder Env sequence , E338K was the sole mutation in the entire 2-A3 Env sequence ( Figure S1 ) . We concluded , then , that this single mutation directly mediated nAb escape . The aspartic acid at position 341 ( D341; HXB2 residue 340 ) , also in the alpha2 helix , changed to asparagine ( D341N ) in one Env ( Figure 2 ) . Finally , the glutamic acid at position 456 ( E456; HXB2 residue 460 ) in V5 switched to lysine ( E456K ) in four Envs ( Figure 2 ) . The potential escape mutations at I295 , D341 , and E456 were introduced into the founder Env 0-B24 by site-directed mutagenesis to determine if these alterations could individually switch the founder Env phenotype from sensitive to resistant when assayed for neutralization by 2-month plasma . In addition , amino acid changes were introduced into escape Env 2-A3 at K338 to determine whether these mutants maintained nAb resistance . The I295R and I295T substitutions in C2 independently conferred nAb escape , with I295R producing a slightly higher level of resistance that was most evident at the 1∶100 dilution of plasma ( Figure 3A ) . For position 338 , two naturally occurring substitutions ( E338D from 2-A23/2-A24 and E338G from 2-B18 , see Figure 2 ) and three artificially introduced mutations ( K338I , K338Q , and K338R ) independently reproduced escape Env 2-A3's wild-type level of resistance , arguing that any change at this position could provide full escape from neutralization ( Figure 3B ) . Thus , the degree of neutralization resistance conferred by changes at I295 , but not at E338 , varied by the amino acid substitution . Introduction of D341N into the alpha2 helix of the founder Env 0-B24 also recapitulated the wild-type resistance level of 2-B12 ( Figure 3C ) . Because the I295 , E338 , and D341 escape mutations occurred independently in the 2-month Envs , each represents a distinct lineage for escape ( Figure 2 ) . In addition , the potency of resistance was substitution-specific; I295R/T and E338D/G/K produced the highest levels of resistance , while D341N lagged somewhat behind and provided partial resistance . In contrast , the E456K mutation in V5 exerted no overt influence on neutralization phenotype when introduced into the founder Env 0-B24 , despite being carried in nearly half of the 2-month escape Envs ( Figure 3D ) . Overall , at 2-months , the viral population utilized a common amino acid substitution mechanism that diverged down three discrete escape pathways , each of which conferred nAb resistance . Though positions 295 , 338 , and 341 appear disparate in the linear gp120 sequence , these residues cluster when plotted onto a 3-dimensional representation of the R880F founder Env sequence , which was modeled using all existing structures for CD4-bound HIV-1 gp120 ( Figure 4 ) . This proposed epitope emerges near the base of the V3 domain , which is well exposed on the native trimer and is also targeted by the broadly neutralizing , glycan-dependent mAb PGT128 [8] . The spatial proximity of these three residues provides evidence for a single nAb epitope during early subtype A HIV-1 infection and an explanation for why a substitution at any one of the three positions independently caused nAb resistance . During HIV-1 infection , the antibodies circulating in patient plasma could ostensibly represent a heterogeneous pool with varying epitope specificities . Although we were able to identify a single , early nAb target in subject R880F using autologous plasma and 3-dimensional modeling , this epitope could be recognized by a polyclonal nAb response mediated by more than one B cell [31] . To illuminate the characteristics of individual monoclonal effectors , we PCR amplified and cloned antibody VH and VL genes from memory B cells present in a cryopreserved R880F peripheral blood mononuclear cell ( PBMC ) sample collected at 16-months post-seroconversion ( Table 1 ) . Multiple VHs and VLs were obtained , but only one VH , named 19 . 3H-HC , neutralized the founder Env when combined with either of two highly related VLs . Sequence analysis revealed that the R880F VH utilized IGHV3-30*02 , IGHD1-7*01 , and IGHJ4*02 gene segments based on matching within the SoDA database [32] and demonstrated 23 . 2% mutation across its framework ( FWR ) and complementarity-determining regions ( CDR ) , as compared with germline at the amino acid level ( Figure 5A ) . The VLs , named 19 . 3H-L1 and 19 . 3H-L3 , were clonal relatives , both using IGLV2-14*01 and IGLJ2*01 gene segments based on matching within the SoDA database [32] and exhibiting mutation rates of 13 . 6% and 14 . 5% from the putative germline , respectively ( Figure 5B ) . Five total amino acid differences between the 19 . 3H-L1 and 19 . 3H-L3 VLs congregated in and around CDR1: 19 . 3H-L3 contained two threonines ( T ) and one phenylalanine ( F ) in CDR1 that were not present in 19 . 3H-L1 , while arginine ( R ) and glutamic acid ( E ) residues arose just downstream of CDR1 in the FWR2 region of 19 . 3H-L1 that were not present in 19 . 3H-L3 ( Figure 5B ) . The VL CDR3 domains of 19 . 3H-L1 and 19 . 3H-L3 were identical and contained five amino acid differences from the putative germline . The two R880F mAbs produced by combination of 19 . 3H-HC and 19 . 3H-L1 or 19 . 3H-L3 are hereafter referenced solely by their VL designations . Figure 5 demonstrates that both 19 . 3H-L1 ( C ) and 19 . 3H-L3 ( D ) neutralized the founder Envs 0-A6 and 0-B24 , although 19 . 3H-L3 did so with approximately one log greater potency . In addition to neutralizing the founder Env , both mAbs neutralized the 2-month plasma escape Env 2-B12 with similar potencies . 19 . 3H-L3 also neutralized plasma escape variants 5-B52 and 2-B31 potently , and 2-A9 and 2-A13 to a much lesser extent . The remaining 2- and 5-month escape variants , and all 7- and 10-month escape variants were resistant to both mAbs . This result suggests that the mAbs are representative of those that circulated within the first few ( 2–5 ) months of infection; because they were isolated from memory B cells , 19 . 3H-L1 and 19 . 3H-L3 do not reflect the ability of the 16-month plasma nAbs to neutralize all longitudinal R880F Envs ( Figure 1F , Table 2 ) . To provide evidence for the specificity and authenticity of 19 . 3H-L1 and 19 . 3H-L3 , the common VH , 19 . 3H-HC , was co-transfected with other autologous VL genes from two randomly selected R880F B cell wells . One VL utilized the same IGLV2-14*01 gene segment as 19 . 3H-L1 and 19 . 3H-L3 ( Figure S2A , E ) ; one did not ( Figure S2B , E ) . Conversely , the 19 . 3H-L3 VL was paired with an autologous VH from a different R880F B cell well ( Figure S2C , E ) . All three chimeric antibody supernatants were assayed for activity against a smaller panel of ten longitudinal R880F Envs , and no neutralizing activity was observed ( Figure S2A–C ) , suggesting that stochastic pairing of R880F VHs and VLs does not confer neutralizing activity . To map the specificity of mAbs 19 . 3H-L1 and 19 . 3H-L3 in finer detail , we utilized the point mutants from Figure 3 , with the addition of double mutant 2-A3 K338G D341N , which was representative of escape Env 5-B52 . As previously mentioned , 19 . 3H-L1 neutralized Env 2-B12 in addition to the founder Envs; Env 2-B12 was the only Env in the panel that shared with the founder Env all three unmutated residues at positions I295 , S335 , and E338 ( Figure 2 , Table 2 ) . A change at any one of these positions resulted in resistance to 19 . 3H-L1 neutralization ( Figure 5C , Table 2 ) . 19 . 3H-L1 neutralized 2-B12 more potently than the founder Envs; this was directly attributed to D341N , as this single substitution introduced into Env 0-B24 ( 0-B24 D341N ) increased the founder Env's sensitivity to that of 2-B12 ( Figure 5C , Table 2 ) . Despite sharing a common VH with 19 . 3H-L1 , 19 . 3H-L3 demonstrated a distinct pattern of specificity . In contrast to 19 . 3H-L1 , 19 . 3H-L3 neutralized the founder and 2-B12 Envs equivalently . In this case , then , the D341N mutation ( 0-B24 D341N ) had very little effect on the neutralization phenotype ( Figure 5D ) . 19 . 3H-L3 also neutralized Envs carrying the I295T substitution ( 0-B24 I295T and 2-B31 ) but displayed a much weaker level of neutralization capacity against Envs containing the I295R substitution ( 0-B24 I295R , 2-A9 , and 2-A13 ) . 19 . 3H-L3 neutralized Envs containing the E338G substitution when it occurred in the presence of D341N ( 5-B52 and 2-A3 K338G D341N ) but not when E338G ( or any other E338 substitution ) occurred in isolation ( Figure 5D , Table 2 ) . R880F mAb 19 . 3H-L3 , therefore , had potent neutralizing activity against two Envs ( 5-B52 and 2-B31 ) and modest activity against two Envs ( 2-A9 and 2-A13 ) that were resistant to contemporaneous plasma and to mAb 19 . 3H-L1 . Hence , the mutational program at positions 295 , 338 , and 341 , first witnessed at 2-months post-seroconversion to facilitate immune evasion ( Figure 3 ) , likely fueled subsequent rounds of nAb recognition , and mutations that originally evolved the virus toward an escaped phenotype here conferred sensitivity to somatically related autologous mAbs ( Figure 5 , Table 2 ) . To ascertain if 19 . 3H-L1 and 19 . 3H-L3 would compete for Env binding , three R880F gp120 monomeric proteins ( the 0-A6/B24 founder Env gp120 , and mutants containing I295R or E338K ) were synthesized , purified , and employed in a competition ELISA assay . To first establish a baseline level of binding , the R880F mAbs were biotinylated and incubated with wild-type 0-A6/B24 gp120 protein . 19 . 3H-L3 demonstrated more robust binding , as compared to 19 . 3H-L1; the negative control mAb 6 . 4C ( directed against a highly specific epitope in V1V2 [31] ) , and the broadly neutralizing mAb PGT128 [8] , which shares epitope space with the R880F mAbs , both failed to bind ( Figure 6A ) . Consistent with the neutralization data in Figure 5 , neither R880F mAb could bind detectably to the I295R or E338K mutant gp120 proteins ( Figure 6B–C ) . Wild-type 0-A6/B24 gp120 protein was then pre-incubated with 19 . 3H-L1 , 19 . 3H-L3 , or the negative control antibody 6 . 4C , washed , and incubated with either biotinylated 19 . 3H-L1 ( Figure 6D ) or 19 . 3H-L3 ( Figure 6E ) to discern if initial pre-incubation could block secondary binding . 19 . 3H-L1 modestly competed with itself ( Figure 6D ) but could not effectively compete for binding with 19 . 3H-L3 ( Figure 6E ) . Conversely , 19 . 3H-L3 strongly competed with both itself ( Figure 6E ) and 19 . 3H-L1 ( Figure 6D ) . Thus , 19 . 3H-L3 neutralizes a greater number of R880F Envs than 19 . 3H-L1 , binds more strongly to the founder 0-A6/B24 gp120 , and neutralizes the Env 0-A6/B24 pseudovirus more potently , underscoring the significance of VL alterations where antigen recognition and neutralization efficacy are concerned . To interrogate the antigen-binding site characteristics of R880F mAbs that influenced their distinct neutralization profiles , crystal structures of the 19 . 3H-L1 and 19 . 3H-L3 Fabs were determined to the resolutions of 1 . 7 Å ( Figure 7A ) and 2 . 7 Å , respectively ( Table S1 ) . Although the two Fabs were crystallized in different space groups , the resultant structures were highly similar , with root mean square deviations less than 1 Å when all of the Cα atoms were superimposed ( data not shown ) . Several structural analyses were employed , including calculations of Optical Docking Area ( ODA , shown in Figure 7B , which predicted the antigen-binding sites by calculating the desolvation free energy of the surfaces ) , surface pockets , and electrostatic surface potentials . ODA analyses indicated that the antigen-binding sites of 19 . 3H-L1 and 19 . 3H-L3 were very flat , forming roughly rectangular shapes approximately 15 Å wide and 30 Å long on top of the six CDR loops ( Figure 7C ) . No pockets existed in these binding surfaces , and the shared CDR H3 , although it was 18 amino acids long ( Kabat numbering scheme [33] ) , did not protrude . Such flat antigen-binding sites likely interact with epitopes formed by residues also on planar surfaces ( i . e . flat-surface antigen-antibody contacts ) . Electrostatic surface potential analyses showed that the 19 . 3H-L1 and 19 . 3H-L3 antigen-binding sites were essentially neutral; a couple of slightly positive regions along one side of the rectangular contact area counterbalanced a slightly negative opposite region ( Figure 7C , blue and red patches , respectively ) . Three CDR1 residues that differed between 19 . 3H-L1 and 19 . 3H-L3 ( Ser/Thr at residue 27 , Gly/Thr at residue 29 , and Tyr/Phe at residue 32; Kabat numbering scheme [33]; Figure 7D ) did not create any substantial structural differences between the two antigen-binding sites . These changes did , however , have the potential to influence antigen-antibody interactions . The Tyr in 19 . 3H-L1 to Phe in 19 . 3H-L3 change at residue 32 likely increased the hydrophobicity at the center of the antigen-binding site , which may have augmented hydrophobic interactions with the antigen . The Gly to Thr mutation at residue 29 added a polar side chain with additional hydrogen binding possibilities . Finally , the Ser to Thr substitution at residue 27 provided a more stable side chain . As a group , these VL alterations probably enhanced the antigen-binding affinity of 19 . 3H-L3 , explaining its increased autologous neutralization breadth . As demonstrated in Figure 5 , D341N appeared to be detrimental to the preservation of a neutralization-resistant phenotype , in the context of mAbs 19 . 3H-L1 and 19 . 3H-L3 during early infection . This mutation was , nonetheless , retained in later escape Envs . Inspection of the 7- and 10-month Env sequences containing D341N revealed that they had acquired additional substitutions , I295N ( HXB2 residue 293 ) and/or S335N ( HXB2 residue 334 ) , absent from earlier Envs ( Figure 2 ) ; each of these mutations affected a potential N-linked glycosylation site ( PNGS ) . Accordingly , we hypothesized that these co-traveling mutations compensated for the vulnerability associated with D341N in a PNGS-dependent manner . To explore this , the I295N substitution , which created a PNGS , was introduced into two mAb-sensitive Envs: 0-A6 and 2-B12 . The I295N versions of these two Envs displayed high-level resistance against mAbs 19 . 3H-L1 and 19 . 3H-L3 ( Figure 8A–B , Table 2 ) . Similarly , the S335N substitution , which also incorporated a PNGS , was inserted in three mAb-sensitive Envs: 0-A6 , 2-B12 , and 5-B52 . The S335N versions of these three Envs also became highly resistant to 19 . 3H-L1 and 19 . 3H-L3 ( Figure 8A–B , Table 2 ) . The S335N substitution shifted a well-conserved PNGS sequon at position 333 ( HXB2 residue 332; Figure 8C ) that is targeted by broadly neutralizing mAbs PGT128 and 2G12 [8] , [34] , [35] . To determine if the observed mAb resistance was glycan-dependent , an S335Q substitution was created in Env 2-B12 . Unlike S335N , which shifted the N333 sequon down two positions , S335Q destroyed the N333 sequon altogether ( Figure 8C ) . The resulting mutant , 2-B12 S335Q , was two logs less sensitive to neutralization by mAb 19 . 3H-L1 than the parental Env 2-B12 , but did not reach the high level of resistance achieved by 2-B12 S335N; in contrast , S335Q had only a slight effect on neutralization by mAb 19 . 3H-L3 ( Figure 8A–B , Table 2 ) . High-level resistance against mAbs 19 . 3H-L1 and 19 . 3H-L3 , therefore , required the addition and/or shifting of PNGS sequons , but amino acid substitution S335Q also provided partial resistance that was much more effective against mAb 19 . 3H-L1 . Together , the data strongly support a mechanism of mAb escape that was PNGS-dependent and may have introduced glycans capable of obscuring the V3-proximal space recognized by 19 . 3H-L1 and 19 . 3H-L3 ( Figure 8D ) . Nevertheless , the two mAbs–common heavy chain notwithstanding–appear to recognize subtly distinct epitopes . The VH , in particular the CDR H3 , has generally been considered a major determinant of epitope recognition and nAb breadth . In our study , VL differences appreciably expanded the neutralization capacity of mAb 19 . 3H-L3 against autologous Envs . To probe whether this increase in breadth carried over to neutralization of heterologous Envs , mAbs 19 . 3H-L1 and 19 . 3H-L3 were tested against a panel of fourteen heterologous Env pseudotypes that included one A/C recombinant , four subtype A , three subtype B , and six subtype C Envs . The mAbs were unable to neutralize any of the heterologous Envs ( Figure 9A–B ) . Thus , while mAb 19 . 3H-L3 possessed increased breadth against autologous Envs as compared to 19 . 3H-L1 , this did not extend to genetically diverse Envs . Regardless of this restricted mAb cross-clade neutralization , R880F plasma collected at 16-months or 3-years post-infection did have similarly moderate breadth against heterologous Envs , which increased in potency over time ( Figure 9C–D ) . An amino acid alignment of Envs from the heterologous breadth panel demonstrated that Envs neutralized with the greatest potency at 3-years post-seroconversion , A-Q461 and C-Z205F ( IC50 values of approximately 1∶1000 ) , contained the N335 ( HXB2 residue 334 ) shifted glycan associated with viral escape from mAbs 19 . 3H-L1 and 19 . 3H-L3 ( Figure 9E ) . Furthermore , Env A-Q461 also incorporated the N295 ( HXB2 residue 293 ) substitution indicative of mAb escape . To investigate if the N295 glycan addition and/or the shifted N335 glycan in R880F Envs could have been partially responsible for the heterologous neutralization capacity that developed in this subject , several glycan knock-out mutants were created and tested with 3-year R880F plasma ( Figure 10A ) . Within A-Q461 , the N295 PNGS was eliminated either alone or in conjunction with the N335 PNGS; the N335 PNGS was also individually knocked out ( Figure 10B ) . The positions of interest were reverted back to the amino acid present in the transmitted/founder Env 0-A6/B24 . For C-Z205F , the N335 PNGS was similarly abolished ( Figure 10B ) . Additionally , two heterologous Envs that were only modestly neutralized but that contained the highlighted glycans , C-Z109F and C-Z214M , were mutated as well . All six of the glycan knock-out mutants exactly mirrored their parental equivalents , suggesting that the particular glycans at positions 295 and 335 did not directly contribute to the breadth observed at 3-years post-infection . These data do suggest , however , that early viral escape events likely influenced how breadth developed in this subject , by expanding what was originally a narrow , regional response at the base of the V3 loop to recognize and neutralize distinct portions of Env across genetically diverse variants .
Several recent studies detail the nAb responses in early subtype B and C HIV-1 infection [24] , [25] , [27] , [29] , [31] , [36] . Here we present the first such study of a subtype A infected individual , R880F , where the initial autologous nAb target was defined , along with the consequent routes of viral escape , and two mAbs from early infection were recovered . The kinetics of autologous nAb induction in R880F generally mimicked those described previously for early HIV-1 infection with subtypes A , B , and C [25]–[27] , [30] , [36] , [37] . Reduced neutralizing activity against contemporaneous Envs at each time point indicated a well-established repeating pattern of de novo neutralization and viral escape in subject R880F . The early escape Env 2-A3 that differed by only one amino acid residue from the founder Envs , 0-A6 and 0-B24 , when combined with a comprehensive panel of mutants , supports the hypothesis that the initial site of nAb recognition was a conformational target at the base of the V3 domain . Specifically , individual mutations at I295 , E338 , or D341 in R880F conferred escape from 2-month plasma antibodies . The region that encompasses these mutations is close to the gp120 surface area targeted by the broadly neutralizing mAb PGT128 ( recovered from a CRF02_AG elite neutralizer ) [6] , [8] , by early plasma nAbs and two mAbs recovered from a subtype B infected seroconverter [25] and by multiple autologous mAbs recovered from two subtype B infected individuals after cessation of antiretroviral treatment [38] . Thus , early nAbs across subtypes commonly target an immunogenic gp120 structure topographically situated near V3 , which is well exposed on the Env trimer . V3-adjacent regions of Env do , nevertheless , elicit strain-specific responses that are easily escaped by multiple pathways . In the study by Bar et al . , nAbs in one of three subjects ( CH40 ) targeted a putative conformational epitope composed of the same regions bordering V3 that we describe here for R880F . CH40 immune evasion in the V3 flanks was , however , preceded by escape mutations in V1; this suggests that this latter region , also immunogenic in early infection , may have been targeted first [25] . Moore et al . recently characterized 2 of 79 subtype C infected subjects who were selected because they developed heterologous plasma neutralization breadth mediated by glycan recognition at HXB2 residue N332 , another V3-proximal position . In each of these individuals , the glycan motif at HXB2 residue N334 was present in the founder Env; N332 evolved later as an escape mutation and was subsequently targeted by nAbs [24] . Interestingly , in R880F , the opposite occurred: N332 ( R880F residue N333 ) was present in the founder Env and shifted to N334 ( R880F residue N335 ) as an escape mutation in some Envs . Furthermore , the development of heterologous breadth in R880F was not facilitated by specific recognition of N334 and , therefore , involved additional determinants and complexity . When juxtaposed , these and our studies underscore how identical mutations , when ordered differently during infection , can sometimes drive divergent phenotypic outcomes . Thus , exposure of B cells to a specific sequence of changes in Env can program the course of nAb breadth . In our previous study of autologous nAb responses during early subtype C HIV-1 infection in subject Z205F , we reported that multiple mAbs targeted the V1V2 domain [31] . These three Z205F mAbs used somatically related IGHV3-15*01 and IGLV2-14*01 germline gene segments and recognized a series of overlapping conformational epitopes centered on residues N134 in V1 and R189 in V2 . Each mAb demonstrated a distinct neutralization profile against early autologous Envs , with variable sensitivity to specific glycans . R880F mAbs similarly utilized a restricted set of IGHV3-30*02 and IGLV2-14*01 germline gene segments , but , in this case , only a single isolated VH exhibited neutralization capacity when paired with the two clonally related VLs named 19 . 3H-L1 and 19 . 3H-L3 ( Figure S2 ) . In a recent study , a single VH was recovered through phage display and conferred neutralization when paired with four somatically related variants of the same kappa VL [39] . Such VL shuffling produced mAbs with varying neutralizing activities , the most potent of which was dependent on one residue in FWR2 and one residue in CDR3 . Moreover , precedent sets of clonally related mAbs that show distinct neutralization potency and/or breadth have been catalogued in HIV-1 infection [6] , [7] , [15] , [38]–[40] . Within the context of our study , it is conceivable that only one R880F VL is authentic , while the other was generated by mutation during short-term in vitro stimulation of B cells . This caveat notwithstanding , variation between the neutralizing activities of mAbs 19 . 3H-L1 and 19 . 3H-L3 highlights a feasible mechanism for gradual acquisition of autologous breadth against highly related escape variants that was directly attributable to VL changes . Furthermore , in future studies it would be advantageous to recover a greater number of distinct antibodies , as our ability to understand breadth fully here was limited with only two highly related mAbs . Notably , the mAbs from Z205F and R880F were predicted to utilize the same VL germline , IGLV2-14*01 . This germline gene segment is also employed by the broadly neutralizing mAbs PG9 and PG16 that target a quaternary epitope involving V1V2 and V3 and is again paired with a VH3 family gene segment , IGHV3-33*05 . These data suggest that VH3 and VL2 pairing is not uncommon for HIV-1 nAbs . Several instances of VH bias for anti-HIV mAbs have been demonstrated based on the epitope: anti-V3 mAbs preferentially use VH5-51 [41] , [42]; anti-CD4i mAbs preferentially use VH1-69 [22]; anti-MPER mAbs in more than one instance also utilize VH1-69 [10]; and anti-CD4bs mAbs preferentially use VH1-46 and VH1-2 [13] , [15] . These pairings may simply reflect common rearrangement of these germline gene segments in the human immunoglobulin repertoire or the structural features that they bind . Defining the structural characteristics of broadly neutralizing mAbs isolated from elite neutralizers in chronic infection has been a major focus in the HIV-1 nAb field . Unlike the Bar et al . study [25] , our data here supply structural information regarding HIV-specific mAbs at the opposite end of the neutralization spectrum . Indeed , we are among the first to report high-resolution crystal Fab structures from early HIV-1 infection , and to show that these mAbs likely mediate planar interactions with antigen that can be subtly altered by VL changes . Structural analyses of the 19 . 3H-L1 and 19 . 3H-L3 antigen-binding sites are consistent with the neutralization data that place their epitopes at the base of the V3 domain . As this region of gp120 lies flat , any one of the three single amino acid changes that conferred escape at 2-months could potentially disrupt the planar interactions between the 19 . 3H-L1 and 19 . 3H-L3 antigen-binding sites and their epitopes , as discussed below . Introducing a positively charged residue with a long side chain ( I295R ) or a glycan ( I295N ) at position 295 is not compatible with the flat hydrophobic surface of the 19 . 3H-L1/19 . 3H-L3 antigen-binding site . In fact , neither mAb could bind to monomeric R880F gp120 containing the I295R mutation . In this model , the I295T substitution would be less effective at conferring neutralization escape . The long , negatively charged E338 side chain is predicted to interact with one of the positively charged surface patches ( Figure 7C , blue ) at the edge of the 19 . 3H-L1/19 . 3H-L3 antigen-binding site , potentially forming a salt bridge with the side chain of a positively charged residue there . The E338K mutation probably destroys this interaction and creates an electrostatic repulsion , which is also consistent with the lack of mAb binding to monomeric R880F gp120 containing the E338K mutation . Interestingly , E338D at this position does not allow 19 . 3H-L1 and 19 . 3H-L3 to neutralize the viruses , suggesting that the length of the Asp side chain is not sufficiently long to restore the possible salt bridge . These results suggest that both length and negative charge of the side chain at E338 are important for antibody binding . The highly conserved N333 ( HXB2 residue 332 ) PNGS at the base of V3 is located at the edge of the proposed epitope and potentially interacts with 19 . 3H-L1 and 19 . 3H-L3 , as removal of this glycan ( S335Q ) weakens the neutralization capacities of these two antibodies , most dramatically in the case of 19 . 3H-L1 . Moreover , the glycan shift from position 333 to 335 ( S335N ) , toward the center of the epitope , also prevents the flat-surface antigen-antibody interaction . In combination , the structural , neutralization , and ELISA binding data indicate that mAbs 19 . 3H-L1 and 19 . 3H-L3 likely recognize overlapping epitopes that are centered on I295 and E338; however , 19 . 3H-L1 is more dependent on D341N and the N333 glycan motif for neutralization than 19 . 3H-L3 . Wholly , these analyses suggest that planar motifs that lie across a flat antigen surface could mediate antibody-antigen recognition in early HIV-1 infection , prior to multiple rounds of viral escape and perhaps more extensive affinity maturation . Additionally , the specific determinants for optimal antigen recognition by each mAb , and the strengths of R880F founder Env gp120 binding , differ slightly as a result of VL variation . In most cases , neutralization breadth in chronic infection has been attributed to the VH , with particular emphasis on the CDR H3 [8] , [22] , [43]–[45] . Few studies have , however , investigated the roots of neutralization breadth , as was done here . We found , somewhat unexpectedly , that in R880F , VL sequence variation influenced mAb 19 . 3H-L3's ability to neutralize two autologous escape variants that were not neutralized by mAb 19 . 3H-L1 during early infection . Significant augmentation of autologous neutralization via minor VL variation ( instead of extensive CDR H3 lengthening ) supports a potential mechanism for how escape variants that differ by only a few amino acids and/or glycans are neutralized . Based on this , we contend that the maintenance of VH-determined epitope specificity while light chain antigen contacts are varied could represent an important breadth-augmenting mechanism for B cells responding to highly related Env escape variants . More dramatic nAb structural adaptations such as the elongation of CDR H3 may require time for development , as longitudinal viral variants establish more complex ploys to escape . Collectively , several factors appeared to shape the antibody maturation pathways in R880F: ( i ) the initial site of nAb recognition , ( ii ) VH and VL rearrangement , pairing , and somatic hypermutation , and ( iii ) repeated exposure to highly related Env escape variants . Our data are consequently consistent with the idea that neutralization breadth arises through the sequential exposure of somatically related B cells to a cascade of viral escape variants presenting altered versions of the same epitope . Additionally , and in contrast to the Moore et al . report [24] , our findings demonstrate that glycans , which arose in response to the initial waves of neutralization , do not always become subsequent targets for later nAbs or promote the potential to develop heterologous breadth . Moving forward , better understanding of how initial immunoglobulin targeting affects downstream neutralization potential could positively impact HIV-1 vaccine design . Our studies suggest that the mere presence of a PNGS does not ensure its recognition by an antibody . Sequential exposure to glycans and other Env variations may be required to drive the type of specialized antibody response associated with elite neutralization . In fact , support for this type of immunization approach has been demonstrated [46] . It is , however , currently unknown exactly how to accelerate somatic hypermutation , lengthening of the CDR H3 , or the acquisition of other adaptations that lead to increased breadth . We propose that a viable vaccination strategy may involve immunizing with a carefully selected series of Env immunogens that mimic defined amino acid and/or PNGS changes that occurred during the natural viral escape process and led to increased neutralization breadth , such as those described here .
Both the Emory University Institutional Review Board and the Rwanda Ethics Committee approved informed consent and human subjects protocols , and subject R880F provided written informed consent for study participation . Longitudinal plasma and PBMC samples were obtained from ART-naïve subject R880F during enrollment in International AIDS Vaccine Initiative ( IAVI ) Protocol C at Projet San Francisco ( PSF ) in Kigali , Rwanda , as part of a multi-site study of early HIV-1 infection in adult Africans . The PSF cohort , which provides voluntary HIV-1 testing , counseling , and condom provision to cohabiting heterosexual couples , is discussed in more comprehensive detail in [47] , [48] . Plasma viral load determination ( reported in Table 1 ) was underwritten by IAVI and performed at Contract Lab Services ( CLS ) in South Africa using an Abbott m2000 system where typical detection ranged between 160 and 4×107 copies/ml . Conditions for plasma viral RNA extraction and purification , cDNA synthesis , and nested single-genome PCR amplification have been described previously [49] . Subsequent full-length Env gp160 coding regions ( plus Rev , Vpu , and partial Nef ) were TA cloned into the CMV promoter-driven expression plasmid pcDNA3 . 1/V5-His-TOPO ( Invitrogen ) and screened for biological function as pseudoviruses following co-transfection with an Env-deficient subtype B proviral plasmid ( pSG3Δenv ) in 293T cells using FuGENE HD ( Roche or Promega ) . Forty-eight hours later , supernatant was collected , clarified at 3 , 000 rpm for 20 min , and used to infect Tzm-bl cells . Following another 48-hour incubation , β-gal staining was performed , and wells were scored positive or negative for blue foci . Fourteen subtype A , B , and C envelopes were used to evaluate the heterologous neutralization breadth of R880F mAbs 19 . 3H-L1 and 19 . 3H-L3 along with autologous 16-month and 3-year plasmas . One A/C recombinant and three subtype C early transmitted variants were previously cloned in our laboratory , as described in [49]: A/C-R66M is R66M 7Mar06 3A9env2; C-Z205F is Z205F 27Mar03 ( “0-month” ) EnvPL6 . 3 [29] , [31]; C-Z1792M is Z1792M 18Dec07 3G7env2; and C-Z185F is Z185F 24Aug02 ( “0-month” ) EnvPB3 . 1 [29] . Ten envelopes were obtained through the AIDS Research and Reference Reagent Program , Division of AIDS , NIAID , NIH: from Dr . Julie Overbaugh , A-Q769-b9 is Q769ENVb9 ( Cat#11545 ) , A-Q769-d22 is Q769ENVd22 ( Cat#10458 ) , A-Q461 is Q461ENVd1 ( Cat#11544 ) , and A-Q23 is Q23ENV17 ( Cat#10455 ) [50]–[52]; from Drs . David C . Montefiori , Feng Gao , and Ming Li , B-SS1196 is SS1196 . 1 ( Cat#11020 ) , and B-TRO is TRO , clone 11 ( Cat#11023 ) [53]; from Drs . Beatrice H . Hahn , Yingying Li , and Jesus F . Salazar-Gonzalez , B-gp160opt is pConBgp160-opt ( Cat#11402 ) , C-gp160opt is pConCgp160-opt ( Cat#11407 ) , and C-Z214M is ZM214M . PL15 ( Cat#11310 ) [54]–[56]; and from Drs . Cynthia A . Derdeyn and Eric Hunter , C-Z109F is ZM109F . PB4 ( Cat#11314 ) [57] . Mutations were created through PCR using two overlapping primers that contained the mutated sequence per Env , in a strategy similar to that described previously [29] , [31] , [58] , [59] . Briefly , the plasmids containing 0-A6 , 0-B24 , 2-A3 , 2-B12 , 5-B52 , A-Q461 , C-Z205F , C-Z109F , or C-Z214M env genes were amplified with the following sets of forward ( F ) and reverse ( R ) primer sequences , where the mutated nucleotides are underlined: For mutants 0-A6 I295N and 2-B12 I295N: F 5′-cagcctgtgaatattacgtgtattagaactggc-3′ and R 5′-gccagttctaatacacgtaatattcacaggctg-3′ For mutant 0-B24 I295R: F 5′-gcccagcctgtgagaattacgtg-3′ and R 5′-cacgtaattctcacaggctgggc-3′ For mutant 0-B24 I295T: F 5′-cttgcccagcctgtgacaattacgtgtattag-3′ and R 5′-ctaatacacgtaattgtcacaggctgggcaag-3′ For mutants 0-A6 S335N and 2-B12 S335N: F 5′-gcatattgtaatgtcaatagaacagaatgg-3′ and R 5′-ccattctgttctattgacattacaatatgc-3′ For mutant 5-B52 S335N: F 5′-gcatattgtaatgtcaatagaacaggatgg-3′ and R 5′-ccatcctgttctattgacattacaatatgc-3′ For mutant 2-B12 S335Q: F 5′-gcatattgtaatgtccaaagaacagaatgg-3′ and R 5′-ccattctgttctttggacattacaatatgc-3′ For mutant 2-A3 K338D: F 5′-gtcagtagaacagactggaatgacactttac-3′ and R 5′-gtaaagtgtcattccagtctgttctactgac-3′ For mutant 2-A3 K338G: F 5′-gtcagtagaacaggatggaatgacactttac-3′ and R 5′-gtaaagtgtcattccatcctgttctactgac-3′ For mutant 2-A3 K338G D341N: F 5′-gtcagtagaacaggatggaatgacactttac-3′ and R 5′-gtaaagtgtcattccatcctgttctactgac-3′ followed by F 5′-gtagaacaggatggaataacactttacaacaggtag-3′ and R 5′-ctacctgttgtaaagtgttattccatcctgttctac-3′ For mutant 2-A3 K338I: F 5′-gtcagtagaacaatatggaatgacactttac-3′ and R 5′-gtaaagtgtcattccatattgttctactgac-3′ For mutant 2-A3 K338Q: F 5′-gtcagtagaacacaatggaatgacactttac-3′ and R 5′-gtaaagtgtcattccattgtgttctactgac-3′ For mutant 2-A3 K338R: F 5′-gtcagtagaacaagatggaatgacactttac-3′ and R 5′-gtaaagtgtcattccatcttgttctactgac-3′ For mutant 0-B24 D341N: F 5′-cagaatggaataacactttacaacagg-3′ and R 5′-cctgttgtaaagtgttattccattctg-3′ For mutant 0-B24 E456K: F 5′-gagatggtggtaaggatattaacag-3′ and R 5′-ctgttaatatccttaccaccatctc-3′ For mutant A-Q461 N295I: F 5′-ccaagcctgtgataattacttgtatcagacctggc-3′ and R 5′-gccaggtctgatacaagtaattatcacaggcttgg-3′ For mutant A-Q461 N335S: F 5′-gcacattgtgttgtcagtagaacagagtggaataac-3′ and R 5′-gttattccactctgttctactgacaacacaatgtgc-3′ For mutant A-Q461 N295I N335S: F 5′-ccaagcctgtgataattacttgtatcagacctggc-3′ and R 5′-gccaggtctgatacaagtaattatcacaggcttgg-3′ followed by F 5′-gcacattgtgttgtcagtagaacagagtggaataac-3′ and R 5′-gttattccactctgttctactgacaacacaatgtgc-3′ For mutant C-Z205F N335S: F 5′-caagcatattgtagcattagtaaaagtaaatggaatgac-3′ and R 5′-gtcattccatttacttttactaatgctacaatatgcttg-3′ For mutant C-Z109F N335S: F 5′-gaaaggcatattgtaaaattagtggaagtgagtggaatg-3′ and R 5′-cattccactcacttccactaattttacaatatgcctttc-3′ For mutant C-Z214M N295I: F 5′-caacttacagaagctgtaataattacgtgtatgaggccc-3′ and R 5′-gggcctcatacacgtaattattacagcttctgtaagttg-3′ The PCR cycling parameters were 1 cycle of 95°C for 2 min; 30 cycles of 95°C for 20 sec , 50°C to 60°C for 20 sec ( the optimal annealing temperature was determined for each primer set ) , and 72°C for 2 min; and 1 cycle of 72°C for 3 min . The samples were then stored at 4°C . The 25-µl PCR mixtures contained 62 . 5–250 ng of each primer , 5–30 ng of the plasmid template , 0 . 4 mM dNTPs , and 1× reaction buffer . PfuUltra II Fusion HS DNA Polymerase ( Stratagene ) was used to generate the PCR amplicons , which were digested with 20 U DpnI ( NEB ) for 1 hr to remove contaminating template DNA and then transformed into maximum-efficiency XL10-Gold ultracompetent cells ( Stratagene ) so that the DNA volume did not exceed 5% of the cell volume . Transformed cells were plated onto LB-ampicillin agar plates , which generally resulted in 50 to 100 colonies . Isolated colonies were inoculated into LB-ampicillin broth for overnight shaking ( 225 rpm at 37°C ) , and plasmids were purified using a QIAprep spin miniprep kit ( Qiagen ) . All env mutations were confirmed by nucleotide sequencing . Sanger DNA sequencing of wild-type and mutant envelope genes and immunoglobulin genes was executed with an ABI 3730xl DNA Analyzer and BigDye Terminator v3 . 1 chemistry at one of two facilities: the University of Alabama at Birmingham Center for AIDS Research ( P30-A127767 ) DNA Sequencing Shared Facility or GenScript . Nucleotide sequences were edited and assembled using Sequencher v5 . 0 and deposited into GenBank under accession numbers JX096639-JX096660 for wild-type env clones and JX124277-JX124282 for immunoglobulin genes . Amino acid sequences were translated and aligned using Geneious v5 . 0 . 3 . Five-fold serial dilutions of heat-inactivated R880F plasma samples , antibody-containing 293T supernatants , or purified R880F monoclonal antibodies were assayed for neutralization potential against viral pseudotypes in the Tzm-bl indicator cell line , with luciferase as the ultimate readout , as described previously [29] , [37] , [57] , [58] , [60] . In short , Tzm-bl cells were plated and cultured overnight in flat-bottomed 96-well plates . Pseudovirus ( 2 , 000 IU ) in DMEM with ∼3 . 5% FBS ( HyClone ) , 40 µg/ml DEAE-dextran was incubated with serial dilutions of plasma , supernatant , or antibody , and subsequently , 100 µl was added to the plated Tzm-bls for a 48 hr infection before being lysed and evaluated for luciferase activity . Data was retrieved from a BioTek Synergy HT multi-mode microplate reader with Gen 5 , v1 . 11 software , the average background luminescence from a series of uninfected wells was subtracted from each experimental well , infectivity curves were generated using GraphPad Prism v4 . 0c where values from experimental wells were compared against a well containing virus only with no test reagent , and IC50 values were determined using linear regression in Microsoft Excel for Mac 2011 , v14 . 0 . 2 . The subject R880F 0-B24 Env gp120 sequence was modeled using the MODELLER program [61] . The template for the homology model was a subtype A gp120 obtained by longtime all-atom molecular dynamics simulation using the CHARMM27 potential in the NAMD program [62] . This simulated gp120 was modeled using all known CD4-bound gp120 structures ( Protein Data Bank [PDB] accession numbers 1G9M [63] , 1RZK [22] , 2B4C [64] , 2NY7 [65] , 3JWD and 3JWO [66] , and 3LMJ [43] ) as templates . In all of these structures , the core of gp120 was highly similar; however , it should be noted that none of these structures is subtype A . Multiple templates were used because it has been shown that this creates high quality homology models . In addition , each template has slightly different regions of gp120 resolved . Before modeling , the templates were arranged in the trimeric state , which has been resolved using cryoelectron microscopy ( PDB accession number 3DNO [67] ) , to ensure that the hypervariable loops did not sterically clash with the neighboring monomers . During modeling , disulfide constraints were added for the conserved cysteines present in all gp120 sequences . All sequence alignments used for modeling templates were based on sequences in the HIV-1 database ( www . hiv . lanl . gov ) . The subject R880F 0-B24 Env gp120 sequence mutated with N295 and N335 was modeled using the protocol described in [31] . Xleap in AmberTools kit 1 . 4 was used to add two glycans at residues N295 and N335 . A five-mannose glycan was used in this simulation because it was found to be the most abundant glycan form in the immunodeficiency virions [68] . Amber99SB force field [69] was used for the gp120 protein , and GLYCAM06 force field [70] was used for the five-mannose glycan . All the systems were minimized using a 3-step protocol in which the protein was gradually allowed to move . These steps were: heavy atoms fixed ( 5 , 000 steps ) , protein backbone atoms fixed ( 5 , 000 steps ) , and all atoms free to move ( 20 , 000 steps ) . The system was gradually heated in a four-step process . The initial temperature was set to 100 K , and only hydrogen atoms were allowed to move for 25 , 000 fs . In the second step , the temperature was set at 300 K , and heavy atoms in the protein were harmonically constrained for the next 25 , 000 fs . Then the temperature was raised to 500 K , and backbone atoms were harmonically constrained for 25 , 000 fs . Force constants for all harmonic constraints were set to 1 kcal mol−1Å−2 . Finally , the temperature was raised to 700 K , and the backbone atoms in the core of gp120 were constrained for the next 4 . 925 ns . The coordinates were saved once every ps , in these 5 ns . The MD simulation was performed using NAMD 2 . 8 [62] . The conformation at the end of the 5 ns MD simulation was used in this study . A viably frozen PBMC sample from subject R880F was collected at 16-months post-infection and was used to recover autologous mAbs 19 . 3H-L1 and 19 . 3H-L3 . The first phase of recovery was performed in the Robinson laboratory . Non-B cells were depleted using immunomagnetic beads ( Miltenyi ) . Approximately 100 , 000 B cells were recovered , and for memory B cell stimulation , the cultures were incubated for 3 days in RPMI medium containing 10% FCS , Epstein-Barr virus , 2 µg/ml R848 ( InVivogen ) , and 100 U/ml IL-2 ( Dr . Maurice Gately , Hoffmann - La Roche Inc . [71] ) . The cells were then plated into 3 , 840 wells in the same medium at low cell densities ( 30 to 50 cells/well ) in forty 96-well tissue culture plates containing irradiated macrophage or human placental fibroblast feeder cells . Starting at 12 days of culture , B cell culture supernatants were screened every 3 to 4 days for antibodies that neutralized the autologous founder pseudovirus containing Env 0-B24 , or that showed ELISA reactivity with 0-B24 Env glycoproteins in previously described assays [31] , [38] . Supernatant from 63 wells screened positive for inhibitory activity in the Tzm-bl assay , and 35 of these were also positive for gp120 binding activity in ELISA . The positive cultures were placed in RNAlater between days 17 to 21 after stimulation . RNA was purified from 12 lysates of these B cell cultures that had been found to be antibody positive . RNA from each well was reversed transcribed into cDNA encoding VH and lambda/kappa VL genes , which were then amplified in a nested PCR as described by Liao et al . [72] . VH and VL gene products were assembled by overlapping PCR into pairs of linear expression vectors encoding full-length human Ig heavy and light chain genes [72] . These vectors were co-transfected into wells containing 80–90% confluent 293T cells . Two days later , supernatants of transfected cultures were tested for antibody activity in the same assays used to screen B cell cultures . One well of 293T cells transfected with VH and VL genes originating from a single B cell culture designated 19 . 3H ( plate 19 , well 3H ) was found to be antibody positive . The antibody ( or antibodies ) produced was thus named 19 . 3H . 293T cells expressing antibody 19 . 3H were serially passaged at limiting cell densities under blasticidin selection ( for maintenance of the Ig vector ) to obtain multiple clones . Selected 19 . 3H-derived clones were expanded into stable antibody producing cell lines to facilitate purification of the antibody by Protein A affinity chromatography . To obtain VH and VL sequences that corresponded to the 19 . 3H antibody activity , VH and VL genes were isolated from the selected 19 . 3H-derived 293T cell clones using two different methods in the second phase of recovery . The first method was used in the Robinson laboratory . VH and VL genes were re-amplified from the selected 293T cell clones and inserted into expression plasmids obtained from InVivoGen: pFUSE-CHIg-hG1 , containing the constant region of the human IgG1 heavy chain , and pFUSE2-CLIg-hl2 containing the constant region of human Ig lambda 2 light chain , respectively . First , the pFuse vectors were linearized by digestion with EcoRI and then subjected to PCR with primers ( IgVH FWD 5′-CGAACCGGTGACGGTGTCGTGGAAC-3′ and REV 5′-ACCGGTGATCTCAGGTAGGCGCC-3′ , IgVLambda FWD 5′-CCAACAAGGCCACACTGGTGTGTCTC-3′ and REV 5′-ACCGGTGATCTCAGGTAGGCGCC-3′ , IgVKappa FWD 5′-GAACTGCCTCTGTTGTGTGCCTGCTG-3′ and REV 5′-ACCGGTGATCTCAGGTAGGCGCC-3′ ) to generate annealing sites of 15 nucleotides that were homologous with ends of the inserts [73] . Second , the SuperScript III One-Step RT-PCR System ( Invitrogen ) was used to amplify the Ig variable regions from 293T-cell-derived mRNA using primers designed to synthesize inserts for use with the ligation-independent In-Fusion cloning system ( Clontech ) . The forward primer IgVH , IgVLambda , IgVKappa FWD 5′-CCTGAGATCACCGGTGCTAGCACCATGGAGACAGACACACTCC-3′ was used for both heavy and light chain inserts and contained a non-annealing tag with 15 nucleotides of homology to the upstream insertion site on the plasmid . Reverse primers for each heavy and light chain ( IgVH REV 5′-CACCGTCACCGGTTCGGGGAAGTAG-3′ , IgVLambda REV 5′GTGTGGCCTTGTTGGCTTGAAGCTCCTC-3′ , IgVKappa REV 5′-CACAACAGAGGCAGTTCCAGATTTCAACTGCTC -3′ ) contained 15–20 nucleotides that overlapped with the 5′ end of the constant regions in linearized pFuse vectors . The In-Fusion reaction was performed according to manufacturer's instructions . Plasmids containing inserts were grown in JM109 competent cells , and at least five colonies were picked for subsequent nucleotide sequencing . A second approach was performed in the Derdeyn laboratory to recover the VH and VL genes from the 19 . 3H-derived 293T cell clones , and from In-Fusion plasmids generated in the Robinson lab , such that all VH and VL genes would be expressed from the same plasmid vector for the neutralization studies . PCR of VH and lambda/kappa VL genes was performed essentially as described by [74] , [75] . Briefly , nested PCR was performed using PfuUltra II Fusion HS DNA Polymerase ( Stratagene ) using the primers described . The first round amplified the leader to constant regions of the VH and VL genes , using cDNA from a 19 . 3H-derived 293T clonal cell line or In-Fusion plasmid DNA as a template . The second round PCR was performed to amplify the variable regions . PCR products were gel purified , digested with appropriate enzymes ( AgeI and SalI for VH , AgeI and XhoI for VL , all enzymes from NEB ) , and cloned into the plasmid expression vectors kindly provided by Dr . Patrick Wilson ( heavy - accession number FJ475055 , lambda - accession number FJ517647 ) . Plasmids were grown in One Shot TOP10 chemically competent E . coli cells ( Invitrogen ) and purified with a QIAprep spin miniprep kit ( Qiagen ) . At least three separate colonies were picked and sequenced . In the end , one VH and two somatically related lambda VL genes were recovered from five 19 . 3H-derived 293T clonal cell lines . The VH combined with either of the VLs ( but not randomly with VLs from other R880F B cell cultures ) produced robust neutralizing activity against the R880F founder Envs 0-A6 and 0-B24 . Further characterization of the mAbs against the larger panel of R880F Envs revealed that the VLs had distinct neutralizing capacities when combined with the 19 . 3H VH , but no neutralizing activity when combined randomly with R880F VHs from other B cell wells . The mAbs containing the different VLs were then designated 19 . 3H-L1 and 19 . 3H-L3 . 293T cells were cultured in T-75 flasks in DMEM with 10% FBS until 80% confluency was reached . Equal amounts ( 6 µg ) of VH- and VL-containing plasmids were mixed with FuGENE HD ( Roche ) at a 1∶3 ratio and used for transfection . After 24 hr , media was removed , cells were washed twice with PBS , and the media was replaced with basal media ( DMEM , 1% PSG , 1% Nutridoma SP ) . Cells were incubated for four days at 37°C , after which the supernatant was harvested . Cell debris was removed by centrifugation at 1 , 500 rpm for 5 min . Approximately 50 ml culture supernatant was used for antibody purification using a Protein A/G Spin column ( Pierce ) according to manufacturer's instructions . Purified antibodies were concentrated using Vivaspin concentrators ( GE ) , and protein concentrations were determined using a Nanodrop spectrophotometer ( BioTek ) . Four monoclonal antibodies were biotinylated with the EZ-Link Sulfo-NHS-LC-Biotinylation Kit ( Thermo Scientific ) for use in ELISA protocols: 19 . 3H-L1 and 19 . 3H-L3 isolated here from R880F , 6 . 4C isolated from Z205F [31] , and PGT128 obtained through the IAVI Neutralizing Antibody Consortium ( NAC ) Protocol G mAb Reagent Program [6] , [76] . For each mAb , 50 µg were diluted in 500 µl 1× PBS ( 0 . 1 M sodium phosphate , 0 . 15 M NaCl , pH 7 . 2 ) for a final protein concentration of 100 µg/ml . A 50-fold molar excess of biotin was incubated with each mAb for 1 hour at room temperature . Excess biotin was removed via Zeba Desalt Spin Column , per the manufacturer's instructions . Reacti-Bind polystyrene 96-well plates ( Thermo Scientific ) were coated overnight at 4°C with 100 µl/well of 2 µg/ml R880F 0-A6/B24 , R880F 0-A6/B24 I295R , or R880F 0-A6/B24 E338K purified gp120 protein ( Life Technologies , GeneArt ) in PBS . Note that blank control wells were coated with gp120 protein but were never subjected to mAb incubation to determine background absorbance , which averaged at 0 . 055 , and assays were performed in duplicate . Plates were subsequently washed six times with 1× PBS-T ( Thermo Scientific; 10 mM sodium phosphate , 0 . 15 M NaCl , 0 . 05% Tween-20 ) and blocked with 200 µl/well of 1× B3T buffer ( 150 mM NaCl , 50 mM Tris-HCl , 1 mM EDTA , 3 . 3% FBS , 2% BSA , 0 . 07% Tween-20 ) for 1 hour at 37°C in a CO2-free incubator . During this incubation step , a two-fold dilution series that spanned 11 wells was prepared in 1× B3T for each biotinylated mAb ( 19 . 3H-L1 , 19 . 3H-L3 , PGT128 , or the negative control , 6 . 4C ) to be tested for binding , beginning at a concentration of 10 µg/ml . Plates were washed six times with 1× PBS-T a second time , and 100 µl/well of serially-diluted mAb was incubated for 1 hour at 37°C . Plates were washed six times with 1× PBS-T a third time , and 100 µl/well of a 1∶10 , 000 dilution of high sensitivity streptavidin horseradish peroxidase ( HRP ) conjugate ( Thermo Scientific ) in 1× B3T was incubated for 1 hour at 37°C . After a final six-time wash with 1× PBS-T , 100 µl of room temperature SureBlue 3 , 3′ , 5 , 5′ tetramethylbenzidine ( TMB ) microwell peroxidase substrate solution ( KPL ) was added to each well and incubated for 5 minutes at room temperature . To cease colorimetric development , 100 µl/well of 2 M H2SO4 was added , and absorbance values at 450 nm were read with a BioTek Synergy HT multi-mode microplate reader . Data was retrieved with KC4 v3 . 4 software , and binding curves were generated using GraphPad Prism v5 . 0d . The gp120 binding ELISA protocol was minimally modified to measure the competitive binding of multiple mAbs , via the following alterations: Only R880F 0-A6/B24 gp120 protein was used , and PGT128 was excluded from the competitions . The first of two 100 µl/well mAb incubation steps was performed via a three-fold dilution series that spanned 7 wells; here , each mAb to be tested for competition ( 19 . 3H-L1 , 19 . 3H-L3 , or the negative control , 6 . 4C ) was prepared in 1× B3T , beginning at a concentration of 10 µg/ml . The second of two 100 µl/well mAb incubation steps involved addition of a constant 1 µg/ml biotinylated competitor ( either 19 . 3H-L1 or 19 . 3H-L3 ) across all wells . Wells were washed six times with 1× PBS-T between these two 1 hour , 37°C incubations . To determine 100% binding for 1 µg/ml biotinylated 19 . 3H-L1 , 19 . 3H-L3 , and 6 . 4C , duplicate wells were incubated with 1× B3T only during the first mAb incubation step and the appropriate biotinylated competitor during the second . The average absorbance for biotinylated 6 . 4C alone was 0 . 056 . Background absorbance averaged at 0 . 048 . Fab fragments of monoclonal antibodies 19 . 3H-L1 and 19 . 3H-L3 were crystallized using previously described methods [41] , [77]–[79] . In short , Fab fragments were generated by papain digestion , purified using affinity and size exclusion chromatography , concentrated , and crystallized with the hanging drop method . Fab 19 . 3H-L1 was crystallized with a well solution containing 0 . 17 M ( NH4 ) 2SO4 , 0 . 085 M cacodylate pH 6 . 5 , 25 . 5% ( w/v ) polyethylene glycol ( PEG ) 8000 , and 15% ( v/v ) glycerol . Fab 19 . 3H-L3 was crystallized with a well solution containing 28% PEG 4K , 0 . 17 M Li2SO4 , 0 . 085 M Tris pH 8 . 5 , and 15% glycerol . X-ray diffraction data were collected at beamline 23-ID-D GM/CA-CAT at the Advanced Photon Source of Argonne National Laboratory , and the data sets were processed using HKL2000 [80] . Crystal structures were solved by the molecular replacement method using MOLREP in CCP4 [81] , [82] . A homologous Fab ( PDB code 3NH7 ) was used as the starting model . The structures were refined using COOT [83] and PHENIX [84] , and analyzed using ICM [85] . The Protein Data Bank ( http://www . rcsb . org/pdb ) accession numbers for the coordinates of the structures of Fabs 19 . 3H-L1 and 19 . 3H-L3 are 4F57 and 4F58 , respectively .
|
Since cases were first recognized in the United States in 1981 , human immunodeficiency virus ( HIV-1 ) has infected over one million Americans . Globally , this scale reaches into the tens of millions , but no effective vaccine exists . Of those infected , approximately 20–30% of patients will develop broadly neutralizing antibodies . The reasons for maturation of these potentially protective responses are presently unknown , but being able to elicit such antibodies via vaccination could curb the pandemic . Here , we defined the earliest neutralizing antibody targets and the consequent routes of viral escape in one subtype A HIV-1 infected subject who developed modest breadth . We also determined the genetic and structural characteristics of early neutralizing monoclonal antibodies circulating in this subject and found that subtle light chain alteration enhanced target contact and neutralization . Overall , our data support the idea that exposure to a specific sequence of viral variants , which have escaped from immune pressure , could program long-term potential for antibody breadth .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"retrovirology",
"and",
"hiv",
"immunopathogenesis",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"infectious",
"diseases",
"viral",
"immune",
"evasion",
"hiv",
"biology",
"viral",
"evolution",
"immune",
"response",
"virology",
"viral",
"diseases",
"immunoglobulins"
] |
2013
|
Viral Escape from Neutralizing Antibodies in Early Subtype A HIV-1 Infection Drives an Increase in Autologous Neutralization Breadth
|
The vocal behavior of infants changes dramatically during early life . Whether or not such a change results from the growth of the body during development—as opposed to solely neural changes—has rarely been investigated . In this study of vocal development in marmoset monkeys , we tested the putative causal relationship between bodily growth and vocal development . During the first two months of life , the spontaneous vocalizations of marmosets undergo ( 1 ) a gradual disappearance of context-inappropriate call types and ( 2 ) an elongation in the duration of context-appropriate contact calls . We hypothesized that both changes are the natural consequences of lung growth and do not require any changes at the neural level . To test this idea , we first present a central pattern generator model of marmoset vocal production to demonstrate that lung growth can affect the temporal and oscillatory dynamics of neural circuits via sensory feedback from the lungs . Lung growth qualitatively shifted vocal behavior in the direction observed in real marmoset monkey vocal development . We then empirically tested this hypothesis by placing the marmoset infants in a helium–oxygen ( heliox ) environment in which air is much lighter . This simulated a reversal in development by decreasing the effort required to respire , thus increasing the respiration rate ( as though the lungs were smaller ) . The heliox manipulation increased the proportions of inappropriate call types and decreased the duration of contact calls , consistent with a brief reversal of vocal development . These results suggest that bodily growth alone can play a major role in shaping the development of vocal behavior .
It is well established ( though often ignored ) that central pattern generators ( CPGs ) are constrained and modulated by the body in which they are embedded [1 , 2] . Moreover , in the field of robotics , much work has demonstrated how the shape and material properties of the body can be exploited to make analogous central control processes simpler; this is known as “morphological computation” [3–5] . In this view , the body is not a device to simply be controlled by the brain , but rather is directly involved in making some behaviors less complicated for the nervous system . In the domain of vocal production , vocal output is typically and reasonably thought to be controlled by a network of CPGs [6–11]; in some species , these CPGs may be activated , modulated , or suppressed by forebrain structures [12–14] ( see [15] for a recent review ) . With regard to morphological computation , studies of birds [16–18] , bats [19] , and humans [20] reveal that the biomechanical properties of the larynx ( or syrinx in birds ) can simplify motor control of vocal production . For example , in zebra finches , discretely different song syllables can be produced by a simple linear driving force exploiting the soft tissue properties of the syrinx [16 , 17] . Along the same lines , simulating the biomechanical properties of the songbird vocal apparatus was also shown to reduce the number of control parameters needed by premotor neurons to organize song structure [12 , 21] . What is not known is the role that morphological computation may or may not play in vocal development . Current investigations of the mechanisms of vocal development typically focus primarily on how changes at the neural circuit level lead to changes in vocal output . For example , the vocal learning literature emphasizes the role played by imitation and the neural changes that may facilitate this behavior , particularly in songbirds and humans [22] . In this case , vocal development is not restricted by body structure , but rather by memory- or motor-related constraints and perceptual predispositions . The possibility of morphological computation is not considered . However , in human infants ( and , logically , all vertebrates ) , there is not only growth in the brain during vocal development , but also growth in the vocal apparatus ( i . e . , the larynx , the vocal tract , and the lungs ) [23–25] . For example , in humans , lung volume nearly triples in size over the first 2 years [26] . Changes in these structures are likely to influence the development of vocal behavior in unexpected ways . Let us illustrate the point from a different domain of behavior: locomotion . A perfect example of morphological computation in development comes from a classic study of human infant stepping behavior [27] . Newborns are able to make well-coordinated stepping movements when held upright , but these movements disappear by the time they reach 2 months of age . While it was assumed by many that the change in stepping behavior was due solely to the developing nervous system ( e . g . , [28] ) , Thelen and colleagues hypothesized that the loss of stepping behavior was due to body growth: the infants’ legs typically fatten up postnatally , and they do not yet have the strength to move heavier legs [27] . To test this hypothesis , they submerged the infants’ legs in water , effectively decreasing their mass . This resulted in the reappearance of stepping and thus falsified the alternative hypothesis that neural change was necessary [27]; the change in behavior was due to changes in the body . We investigated whether developmental changes in vocal production are in part the result of morphological computation using marmoset monkeys as a model system . When out of visual contact of conspecifics ( undirected context ) , adult marmoset monkeys exclusively produce and exchange contact “phee” calls [29 , 30] . Infant marmosets , however , produce mature and immature versions of this contact calls as well as calls that are inappropriate for the undirected context: trills and twitters [31 , 32] . Thus , in the undirected context , the goal of marmoset vocal development is to produce solely contact calls [31 , 32] . The infant trills , twitters , and contact calls have distinct spectral and temporal profiles ( Fig 1A ) [31 , 32] . Call duration , for example , readily distinguishes the syllables of trills , twitters , and contact calls ( Fig 1B ) . The production of the longer duration contact calls is energetically costly ( relative to trills and twitters ) , requiring sustained respiratory power [31 , 33 , 34] . Over the course of development , contact calls gradually increase in duration , becoming more adult-like; they increase in proportion as well [31] ( Fig 1C and 1D ) . The short-duration trills and twitters , however , simply disappear over time in the undirected context ( Fig 1E and 1F ) . Given the energetics required to produce contact calls , we hypothesized that increases in lung capacity via body growth—without any developmental changes in the neural properties of the vocalization-related CPGs—can explain the disappearance of short-duration trill and twitter calls . As the lungs get bigger , the respiration rate slows down because inspiration and expiration take longer [35]; sensory feedback from the lungs to vocalization-related CPGs mediates this influence on vocal output [36] . Although there is also sensory feedback from the larynx [36] , the vocal developmental data show much more pronounced changes in lung capacity–related duration of calls ( Fig 1G and 1H ) than in their laryngeal-related fundamental frequencies ( Fig 1I and 1J ) . We believe this slowing of the respiration rate results in the disappearance of trills and twitters while increasing the proportion and duration of contact calls . To explore this possibility , we generated a numerical model of infant marmoset monkey vocal development and then tested model predictions by placing infant marmosets in a heliox environment and recording its effects on vocal production . The heliox environment simulates a developmental reversal of lung growth by increasing the rate of respiration . Our data show that the decreasing numbers of context-incorrect trills and twitter calls , and the increasing number and duration of the context-appropriate contact calls , are driven by morphological growth and not necessarily developmental changes in the intrinsic activity of neurons .
Let us provide the background on which our model is based . In previous studies , we showed that there is a slow 0 . 1 Hz oscillatory pattern in the spectral entropy and duration of vocalizations as marmoset infants produce highly stochastic vocal sequences [32 , 37] . In other words , there are alternating patterns of noisy ( broad-band ) and tonal calls , as well as long and short duration calls in these infant vocal sequences . These findings suggested that this complex vocal output is governed by neural dynamics that are under a strong influence of a low dimensional input that also exhibits such slow oscillations . We found that arousal levels—as measured by heart rate changes—accounted for this 0 . 1 Hz oscillatory pattern in vocal acoustics [32] . These findings translate into the following scenario: ( 1 ) when the animal is at its lowest or highest arousal levels , it produces immature ( highest entropy ) or mature ( lowest entropy ) sounding contact calls , respectively; ( 2 ) the shorter calls , twitters and trills , are generated between those two states . We thus sought a nonlinear model that can bridge the fluctuating arousal levels to the complex vocal behavior . We built the model with the minimal set of control parameters that can generate infant marmoset monkey vocalizations . These consist of the subglottal pressure and the laryngeal tension [31 , 34] . This model provides the dynamics of the CPGs governing those two parameters . As the respiratory pressure energizes the vibration of vocal folds during the expiratory phase , it determines the duration of phonation . At certain pressure levels , laryngeal tension determines the fundamental frequency of the sound . In order to produce a long duration contact call , a long expiration is needed , as well as a constant laryngeal tension . In other words , the respiratory CPG has to oscillate at a slow rate while the laryngeal CPG is at a stable fixed point . To produce a trill call , respiration needs to provide a relatively sustained pressure while the laryngeal CPG provides a fast oscillating input . By contrast , a twitter call requires a strong , fast oscillation in respiration to break the sound into short syllables; this can be realized by coupling with the fast laryngeal oscillator . We thus set up our model with the topology that two distinct regions of stable fixed points are separated by an oscillatory regime ( see Methods for details ) . The model is composed of two coupled oscillators with distinct natural frequencies: one CPG for respiration and the other for governing the oscillating tension of laryngeal muscles ( Fig 2A ) [38] . The activity levels of the CPGs are modeled by the amplitude of the oscillators . Both CPGs receive a common drive representing the arousal levels . This is similar to the linear drive used to generate nonlinear shifts in gait dynamics in spinal cord CPG models of locomotion [39] . In our autonomous model , the arousal input is the only time-dependent variable that drives the system across different dynamical regions . These CPG dynamics are then converted into air pressure and tension , which are fed into a biomechanical model of the marmoset monkey vocal apparatus [31 , 34] . One assumption key to the model is that the oscillations of the CPGs are adapted to the mass of the lungs [40 , 41] . It has been shown independently in a study of birdsong that the respiratory patterns can be generated in a Wilson–Cowan neural network integrated with the lung ( air sac ) dynamics via sensory feedback [42] . We also implemented a Wilson–Cowan model and found that the period of the oscillations is positively correlated with the mass of the organ ( S1 Fig ) . We used this to justify setting up our main CPG model so that the CPG driving the lungs is much slower than the one driving the larynx . Furthermore , we implemented the effects of changing lung mass on the frequency of the oscillator by varying its time constant , i . e . , greater lung mass yields slower oscillations . In the initial model representing a postnatal day 1 marmoset infant ( with a small lung size/damping coefficient ) , all marmoset call types can be generated by solely and linearly increasing the drive over the course of 6 s . Fig 2B shows that contact calls , trills , and twitters are generated ( s , time-amplitude waveforms and f , spectrograms ) . This 6-s drive I is consistent with the ramping up phase of arousal fluctuations that underlie the production of real infant marmoset vocalizations [32] . The respiratory patterns generated by the model are qualitatively similar to the electromyography ( EMG ) recordings of marmoset infant respiratory patterns ( Fig 2C ) [32] . Low and high drive levels produce relatively constant CPG control of the laryngeal tension , k . This results in the spectrally flat contact calls ( Fig 2D , left panels ) . Moderate drive causes the laryngeal CPG to oscillate around limit cycles , yielding trills and twitters ( Fig 2D , right panels ) . The dynamics of the respiratory CPG are modulated by the laryngeal CPG via the coupling term . Depending on the oscillatory amplitude of the laryngeal CPG relative to the respiratory CPG , respiration can be shortened ( trills ) ( Fig 2B; at 2 s and 4 s of respiration ) or broken into minibreaths to produce twitters ( minibreaths are the small respiratory oscillations added on top of a DC level [43]; Fig 2B; at approximately 2–4 s of respiration ) . We can now use the model to test the hypothesis that lung growth alone can account for the decreasing numbers of trills and twitters ( Fig 1D and 1E ) and increasing number and duration of contact calls ( Fig 1C and 1F; see S1 Text for details ) . Fig 3A and 3B show that if we increase lung size , then we change the patterns of vocalization-related respiration . In this scenario , short duration respiratory patterns decrease in number ( Fig 3B ) . Because the lung capacity scales proportionally with body mass [44] and body mass increases almost linearly with time over the first two months in marmosets ( n = 13; Fig 3C ) , we fit the damping coefficient as a linear function of postnatal days to simulate a biologically plausible trajectory of lung growth . Fig 3D shows the proportion of call types as a function of the drive ( I ) level and increasing lung size . As the lungs grow , the range of I that generates trills and twitters decreases . The pattern of declining trills and twitters generated by the model is similar to the developmental pattern exhibited by real infant marmosets ( Fig 3E and 3F ) , as is the increasing number of contact calls ( Fig 3G and 3H ) . The best fit between the model and data was R2 = 0 . 77 . Our model shows that sensory feedback from growing lungs—without any changes to the CPGs themselves—can account for the decreasing proportion of trills and twitters . Thus , morphological computation is a plausible mechanism for vocal development in this case . To empirically test the model’s predictions , we manipulated the physical property of respiration by placing the developing marmoset infants in a helium-oxygen ( heliox ) environment . Because the air is lighter in a heliox environment , less time is needed to complete a respiratory cycle when the same amount of force is provided by the respiratory muscles ( see S1 Text ) [45] . An increase in the infant marmosets’ respiratory rate in the heliox environment would simulate the temporal dynamics of smaller lungs and allow us to test the predictions of lung growth on vocal output . We did this approximately every other day , from P1 to P60 ( n = 3 subjects , n = 19 , 19 , and 27 sessions ) . For each session , we recorded vocalizations for 10 min in heliox and 10 min in air . The order of these two conditions was counterbalanced . To allow for gas concentration to stabilize when transitioning between heliox and air , we only analyzed the vocalizations in the last 5 min of each 10-min interval . For two infant marmosets , we measured the heliox-induced change in respiration rate via video analysis of abdominal movements while they produced contact calls simply to confirm the obvious: that respiration rate should increase in the heliox . Fig 4B shows averaged traces of respiration of two infants in heliox ( n = 37 traces ) and in air ( n = 25 traces ) , demonstrating that there is an increase in the rate of respiration during the production of calls . Fig 4C shows the mean increase in the respiration rate across the two infants ( 24 . 9% ± 3 . 6% increase , mean ± SEM; p = 4 . 0×10−8 , unpaired 2-tailed t test ) . As predicted by the model , placing infant marmosets in heliox increased the proportion of trills and twitters and decreased the duration of contact calls , a reversal of the vocal development trend ( Fig 4D ) . Fig 4E , 4G and 4I show the developmental trajectories of these three call types produced in and out of the heliox environment for all three infants . For all of them , more trills and twitters are produced , and the duration of contact calls shortened , in heliox . Fig 4F , 4H and 4J show the mean change between heliox and air over the first two months . The proportion of trills increased significantly in heliox ( 36 . 5% ± 6 . 0% increase , mean ± SEM; p = 1 . 45×10−9 , effect size = 0 . 134 , generalized linear model [GLM] ) , as did the proportion of twitters ( 41 . 8% ± 5 . 3% increase , mean ± SEM; p = 1 . 80×10−15 , effect size = 0 . 266 , GLM ) . Contact call duration under heliox condition dropped by 11 . 0% ± 0 . 2% compared to those produced in air ( mean ± SEM; p = 0 , effect size f2 = 0 . 13 with power = 1 . 0 , GLM ) . To rule out the possibility that the heliox manipulation affected the animal’s arousal levels and thus consequently caused the observed differences in vocal output , we performed pairwise comparison on the call rate ( number of calls produced per min ) between these two conditions . Increased call rates are typically associated with increased arousal levels [46] . We did not observe differences in the amount of calls produced as function of heliox versus air ( p = 0 . 65 , Wilcoxon signed rank test ) . These data demonstrate that developmental changes in lung capacity can account for the changing call types produced as the infant marmoset grows . An additional possibility is that the heliox makes it easier to produce trills and twitters via a laryngeal influence , as it is well known that heliox can affect the spectral properties of vocalizations [47–49] . When compared to vocalizations produced in air , heliox shifts the resonant frequency of the vocal tract ( the oral and nasal cavities ) [50] , enhancing the second harmonic of the vocalizations’ spectra ( S1 Text ) . However , heliox does not have a large effect on the fundamental frequency ( F0 ) [47–49] , and the F0 represents the source sound coming directly from the larynx [50] . We calculated the mean power spectral density ( PSD ) of contact calls , trills , and twitters across all postnatal days in both heliox and air . All three call types had nearly identical F0s ( Fig 4K , left panels ) . Heliox significantly enhanced the second to first harmonic amplitude ratio of all tonal calls by approximately 18 dB ( p = 1 . 0×10−63 , effect size d = 0 . 23 , unpaired 2-tailed t test ) . In contrast , the F0s of the tonal calls were increased by only approximately 1 . 8% in heliox ( p = 5 . 4×10−27 , unpaired 2-tailed t test , however with a very small effect size d = 0 . 14 ) . Thus , the heliox effect on the spectral properties of vocalizations was mostly passive and only minimally due to changes in the effort for laryngeal control . That the heliox environment had largely the same effect on all three call types suggests that air density does not differentially benefit the production of trills and twitters .
Vocal development is a consequence of many interacting factors , including the growth of the vocal apparatus , the muscles that innervate it , the nervous system that controls those muscles , and social interactions that adjust nervous system function via experience [34 , 51] . Previous efforts isolated the role of social interactions on marmoset monkey vocal development [31 , 52 , 53] . Those studies took advantage of individual differences in the amount of social feedback provided by parents while controlling for the contributions of body growth . They found that the rate of developmental changes in some acoustic parameters , such as the noisiness and amplitude modulation , could be attributed to the amount of social feedback provided by parents , while other parameters , such as duration , dominant frequency , and the disappearance of calls produced in the incorrect context , could not be explained with such experience-dependent mechanisms [31 , 52] . Conversely , in the current work , we found that increases in call duration and changes in call usage ( i . e . , the disappearance of calls produced in the incorrect context ) could be attributed solely to the growth of one part of the vocal apparatus—the lungs , which provide the respiratory power to produce vocalizations . Our model of interconnected laryngeal and respiratory CPGs predicted that if the respiratory CPG received sensory feedback from growing lungs , the production of trills and twitters would decrease and the production and duration of contact calls would increase . No changes in the neural properties of the CPGs were required . These predictions were empirically tested by recording the vocalizations of infant marmosets in a heliox environment . Akin to placing infants in water to reduce the load on stepping behavior [27] , placing vocalizing infant marmoset monkeys in heliox reduces the respiratory load , thereby increasing the number of trills and twitter calls and shortening the contact calls . Thus , in contrast to the strong emphasis on neural changes typically used to explain vocal development [22] , these data support the idea that some aspects of vocal development can occur through morphological computation: The body ( in this case , the growing lungs ) can be exploited as a computational resource by reducing the number of control parameters that need to be tracked and adjusted by the nervous system [3–5] . Respiration plays a key role in both vocal production [8 , 54] and behavior in general [55] . Respiration and locomotion , for instance , are synchronized to different extents depending on the mechanical constraints imposed by posture and body size on respiration [56] . The upright posture of humans reduces the influence of gait on respiration , allowing more flexibility in respiratory patterning [56] . Our model proposed that morphological computation through lung growth benefited the neural control of vocalization by weakening the coupling of respiration from laryngeal movements . The apparent decoupling of respiration from laryngeal influence by lung growth in marmosets may in the same way allow more independent control of respiration , thereby improving the accuracy of vocal communication . Naturally , our model is a simplification of what is known about the dense , multinode network of vocalization-related CPGs [10] , which includes within it a complicated network of respiratory CPGs [57] . Nevertheless , our model and behavioral data provide supportive evidence for the hypothesis that the intrinsic properties and connectivity of these networks need not change over the course of vocal development to account for some dramatic shifts in vocal output . It is important to note that the model presented in this work provides only one possible solution to the structure of the neural activity that can generate sequences of marmoset infant vocalizations . We did not design the model to simulate any specific neuroanatomical or neurophysiological details , as these are not yet well understood . Rather , we use the model as a way to extract a low-dimensional representation of this complex vocal behavior . There are other dynamical models with similar structures that can also lead to the same results . For example , an alternative setup of the CPG model would be one with articulate CPGs driving different neural populations that , respectively , drive the laryngeal and respiratory muscles . Although our study suggested that the bifurcations that create different vocal patterns occur at the level of subcortical CPGs , it is not sufficient to refute the alternative possibility that forebrain structures might also play an important role [14 , 58 , 59] . Given that vocal development consists of a number of “moving parts” in the body and the brain , we need to understand how these parts and their relationships change over time to produce mature vocal behavior [34] . This integrative understanding is important from a clinical perspective as well . Human infants who do not vocalize a lot tend to be fed and held less by mothers , and are slowed in their speech development [60] . The lack of adequate early vocal output by infants may be due to many factors , including problems related to nervous system function such as arousal dysregulation or motor control deficits , weak laryngeal and respiratory muscles , and/or abnormal growth of the vocal apparatus: the larynx , orofacial cavity , and lungs . It is important that one considers the “whole system” when trying to understand how any behavior works or may go awry .
All experiments complied with the Public Health Service Policy on Humane Care and Use of Laboratory Animals and were approved by the Princeton University Institute Animal Care and Use Committee ( protocol number 1908–15 ) . The vocal development trajectory was constructed partially from a subset of previously published dataset ( n = 10 subjects ) [31] and partially from the control condition of the three subjects used in this work . The subjects are infant common marmosets ( Callithrix jacchus ) housed at Princeton University . The colony room is maintained at approximately 27°C and 50%–60% relative humidity , with 12L:12D light cycle . The subjects were all born in captivity and raised by family . All subjects , including all other members in the family , received water ad libitum and were fed with standard commercial chow supplemented with fruits and vegetables . All experiments were approved by the Princeton University Institute Animal Care and Use Committee . In the search for an appropriate model , we considered a two-dimensional system for each CPG oscillator to allow Hopf bifurcations . We also looked for a model that contains two regions of stable fixed points separated by a limit cycle region via Hopf bifurcations . For simplicity , we start building the dynamical model for each oscillator from a simple 2D system {x˙=yy˙=f ( x , y , a ) , in which f ( x , y , a ) is a polynomial up to the third order and a is a parameter . To allow the location of the fixed point to change from low to high values as the parameter varies monotonically , we simply let x* = a be the only fixed-point solution in our model . Thus , f ( x , y , a ) can have the form f ( x , y , a ) = σ ( x−a ) + g ( x , y ) y , where σ is a constant and g ( x , y ) is a polynomial up to the second order . With these simplifications , the Jacobian matrix has the form J= ( 01σg ( a , 0 ) ) . To have Hopf bifurcations occurring twice , σ < 0 and g ( a , 0 ) switches signs twice; thus , it can be a parabola passing 0 twice . In addition , to allow oscillations in the middle range of a , the parabola is inverted . Hence , g ( x , y ) can have the form g ( x , y ) = μ ( b – x2 ) + h ( y ) , where μ < 0 and b > 0 are constants and h ( y ) is a polynomial with the lowest order of one and highest order of two . Again , we drop h ( y ) for simplicity . Without losing generality , we let σ = −1 , μ = −1 , and b = 1 . We also introduce a coupling term from the other oscillator in the equation and a time constant to change the oscillating frequency . The complete model is as follows {x˙i=yiy˙i=γi2ai−γi2xi+γiyi−γixi2yi+γi2κ ( xj , yj ) , in which ai is the drive input to oscillator i , γi is the time constant for oscillator i , and κ ( xj , yj ) is the coupling input from oscillator j . We assume that the coupling is linear and let κ = αjiyj , in which αji is the coupling strength . Greater γi corresponds to faster oscillation . We define I=a−aminamax−amin∈[0 , 1] for the relative drive strength . The parameters of the model are listed in Table 1 . The sensitivity of the model’s dependence on the parameters is analyzed in S3 Fig . We saturated x1 and x2 with sigmoid functions to get biologically reasonable p ( air pressure , varying between–p0 and p0 ) and k ( laryngeal tension , varying between 0 and k0 ) : {p ( t ) =p0tanh ( x1 ) , k ( t ) =k01+e− ( x2−x20 ) . The behavior of the CPG dynamics was visualized using the phase portraits , in which the oscillatory amplitude xi was plotted against the velocity yi = dxi/dt . With different values of the input , different dynamics were produced . We estimated the proportion of different call types using the bifurcation diagram of x1 in the parameter space of I and γ1 . As different call types are characterized by different duration ( Fig 1B ) , we used the spectrum of x1 to find the regimes for different calls . For each combination of I and γ1 , we iterated in solving the ODE using the Runge–Kutta method 2 , 000 times ( after we discarded the first 500 iterations ) with 0 . 01 step size . The regions for different call types were identified based on the oscillatory frequencies . Call proportions were estimated as the range for call type i in the [0 , 1] range . To compare the model with real data , we found the parameters β0 and β1 for the linear transform PND = β0 + β1γ1 that led to the least sum of squares β0 , β1=argminβ0 , β1∑i=12∑j=1N ( pi ( γ1j ) −p˜i ( β0+β1γ1j ) ) 2 , where pi and p˜i are the simulated and real proportion of call type i ( twitter and trill ) . To estimate the goodness of fit , we calculated the R2 between data and simulated proportions . Starting from P1 , marmoset infants were placed in an induction chamber that holds approximately 45 L of air . The subjects were introduced into the chamber through the lid on top of the chamber . Heliox ( 20% oxygen and 80% helium ) was passed through the inlet on the chamber and air was expelled from the outlet ( Fig 3A ) . An air flow meter was attached to the inlet . A microphone ( Sennheiser MKH 416-P48 ) was placed inside the chamber to record vocalizations . To reduce echoes , acoustic foam was attached to the walls of the chamber . An oxygen sensor ( PASPORT Oxygen Gas Sensor-PS-2126A ) was placed inside the chamber to monitor oxygen concentration throughout the experiment . In the control condition , we replaced the solid lid with a perforated lid . In each session , we carried out recordings of 10 min in heliox and 10 min in air . The order of these two conditions alternated every session . Since it requires 5 min for the gas to fill up the chamber , we discarded the first 5-min recording in both heliox and air conditions in the analysis . Heliox was provided constantly through the heliox session . To control the auditory effect from the heliox injection , we recorded the sound of airflow in the chamber and played it through a Bluetooth speaker ( Lyrix Jive Jumbo ) placed in the chamber near the inlet during the control condition . The sound pressure level of the playback was calibrated the same as the actual airflow sound using a sound level meter ( Extech 407730 ) . An HD webcam ( Logitech C930e ) was placed in front of the chamber facing the side where there was no foam attached to record the abdominal movement during vocalization at 30 fps . To test if the heliox approach was effective , we extracted respiratory pattern of the abdominal movement duration vocalization from video recordings ( Logitech C930e ) . Phonation requires about 5- to 30-fold of pressure more than the baseline breathing , and therefore , it depends upon the abdominal sheet to drive active expiration during vocalization [61 , 62] . We extracted abdominal movements from video recordings in two marmosets who were approximately 2 months old during the production of phee calls in air and heliox environments . Infants at this age essentially only produce phee calls in isolation . Movie clips during phee call production were segmented using Windows Movie Maker . The marmosets usually stay still during vocalization , and so we could select a rectangular area around the abdomen through the frame stack and track its movements during the time window of vocalization . The RGB images were converted to grayscale by taking the mean across the color dimensions . The areas were first vectorized and converted into a matrix with rows representing frames and columns representing pixels . Principle component analysis was carried out to capture frame-to-frame variations related to respiration . The principle components were then aligned with the sound signal , and the PCs that were correlated with vocal production were selected to represent the abdominal movements during vocalization ( S2A Fig ) . The average traces of the video extractions were calculated from the resampled data at 100 Hz and were aligned to the call onsets . We averaged the traces of the same condition . To compare the respiratory rates in different conditions , we calculated number of cycles per s using number of cycles divided by total call duration . To justify this method , we also compared the results with EMG recording ( S2B Fig ) . Onsets and offsets of individual utterances were automatically detected using a custom-made MATLAB routine . Call types were first categorized automatically based on duration and Wiener entropy and then manually inspected . Duration was calculated as the duration of individual utterances within a call . Consecutive utterances in the same category with no more than 0 . 5-s gaps were grouped as one call . Each point of the call type proportions was calculated by grouping two consecutive , counterbalanced sessions . Call proportions were calculated as the number of calls of a specific type divided by the total number of calls in this condition . The corresponding postnatal days were calculated as the mean of the two consecutive days . The PSD of the vocalizations ( per syllable ) was estimated using Welch’s method by applying the MATLAB pwelch function . The F0 was identified as the first peak of the sound spectrum . The second harmonic ( F1 ) was identified as the second peak of the spectrum . The amplitude ratio between F1 and F0 was calculated as the ratio of the mean amplitudes at F1 and F0 within a syllable . We used MATLAB csaps function to fit the data over the first 60 postnatal days for individuals . The 95% confidence intervals were constructed by randomly sampling the data with replacement 1 , 000 times and fitting cubic spline using csaps for each bootstrap sample . MATLAB fitglm routine was used to fit the GLM to the occurrences of trill or twitter over the first two months in all three subjects . In this model , we tested the effect of heliox condition and also controlled for individual differences . We assumed that the response variable follows binomial distribution , and in Fig 4G , we fitted a multiple logistic regression model for the occurrences of trill logit ( Itrill ) =α+β1*S2+β2*S3+β3*Iheliox+ϵ , where S2 and S3 are dummy variables for subject #2 and #3 encoded as S1 = 00 , S2 = 01 and S3 = 10 , Iheliox = 0 or 1 for air condition and heliox condition and ϵ as the random error . Similarly , in Fig 4I , we fitted the model logit ( Itwitter ) =α+β1*S2+β2*S3+β3*Iheliox+ϵ . We used the fitted β3 and its standard error to estimate the mean difference in proportion between the two conditions with subject difference taken into account . The significance of the heliox effect was accessed by the p-value of β3 from the fitglm output . To estimate the effect size of the GLM , we compared the areas under the receiver operating characteristic ( ROC ) curves for a model with the condition variable included and one without it [63] . The area under the curve ( AUC ) was calculated using the MATLAB routine perfcurve , and the ratio of AUC was calculated as r=AUCc−AUC0AUC0−0 . 5 , where AUCc is the AUC with the condition variable and AUC0 is the one without that variable . As in practice we compared the area above the diagonal line , we subtracted 0 . 5 in the denominator . To evaluate the heliox effect on duration , we calculated the duration of the contact call syllables under each condition as a fraction of the daily mean duration in air condition . We assumed that the fractional duration is normally distributed and fitted a general linear model to the fractional duration as a function of subject identity #2 and #3 and condition d=α+β1*S2+β2*S3+β3*Iheliox+ϵ . β3 was then used to estimate the reduction of syllable duration . The effect size was estimated using Cohen’s f2 method for multiple regressions . Power analysis was carried out using the G*Power 3 . The spectral features , F0 and amplitude ratio of F1/F0 , were compared between the two conditions using unpaired 2-tailed t test . Effect size d estimation and power analysis were performed in G*Power 3 .
|
In robotics , the shape and material properties of the robot body can be exploited to make central control processes simpler; this is known as “morphological computation . ” In this view , the body is not a device to simply be controlled by the brain , but rather is directly involved in making some behaviors less complicated for the nervous system . We tested this idea in a real biological system by investigating how marmoset monkey infants change their vocal behavior over time . It would typically ( and reasonably ) be presumed that changes in vocal production are the result of learning and , thus , changes in the nervous system . However , using a computational model , we show that one major feature of developing vocal behavior—the decline in the production of context-inappropriate vocalizations—could simply be the result of lung growth ( a change in body morphology ) without any concomitant changes in central nervous system structure . We then tested the model predictions by placing the infants in a helium–oxygen ( heliox ) environment . This , in effect , simulated a reversal in lung growth and , as predicted , resulted in a reversion back to immature vocal behavior . Thus , morphological computation plays a role in vocal development . These data underscore the importance of considering the whole organism , not just the nervous system , when trying to understand how any behavior works or may go awry .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"syllables",
"medicine",
"and",
"health",
"sciences",
"vocalization",
"linguistics",
"nervous",
"system",
"vertebrates",
"marmosets",
"social",
"sciences",
"animals",
"mammals",
"animal",
"signaling",
"and",
"communication",
"lungs",
"animal",
"models",
"primates",
"respiratory",
"system",
"animal",
"behavior",
"experimental",
"organism",
"systems",
"zoology",
"respiratory",
"physiology",
"research",
"and",
"analysis",
"methods",
"monkeys",
"new",
"world",
"monkeys",
"behavior",
"grammar",
"eukaryota",
"anatomy",
"physiology",
"phonology",
"biology",
"and",
"life",
"sciences",
"amniotes",
"organisms"
] |
2018
|
Vocal development through morphological computation
|
Multiple sclerosis ( MS ) is a complex trait in which allelic variation in the MHC class II region exerts the single strongest effect on genetic risk . Epidemiological data in MS provide strong evidence that environmental factors act at a population level to influence the unusual geographical distribution of this disease . Growing evidence implicates sunlight or vitamin D as a key environmental factor in aetiology . We hypothesised that this environmental candidate might interact with inherited factors and sought responsive regulatory elements in the MHC class II region . Sequence analysis localised a single MHC vitamin D response element ( VDRE ) to the promoter region of HLA-DRB1 . Sequencing of this promoter in greater than 1 , 000 chromosomes from HLA-DRB1 homozygotes showed absolute conservation of this putative VDRE on HLA-DRB1*15 haplotypes . In contrast , there was striking variation among non–MS-associated haplotypes . Electrophoretic mobility shift assays showed specific recruitment of vitamin D receptor to the VDRE in the HLA-DRB1*15 promoter , confirmed by chromatin immunoprecipitation experiments using lymphoblastoid cells homozygous for HLA-DRB1*15 . Transient transfection using a luciferase reporter assay showed a functional role for this VDRE . B cells transiently transfected with the HLA-DRB1*15 gene promoter showed increased expression on stimulation with 1 , 25-dihydroxyvitamin D3 ( P = 0 . 002 ) that was lost both on deletion of the VDRE or with the homologous “VDRE” sequence found in non–MS-associated HLA-DRB1 haplotypes . Flow cytometric analysis showed a specific increase in the cell surface expression of HLA-DRB1 upon addition of vitamin D only in HLA-DRB1*15 bearing lymphoblastoid cells . This study further implicates vitamin D as a strong environmental candidate in MS by demonstrating direct functional interaction with the major locus determining genetic susceptibility . These findings support a connection between the main epidemiological and genetic features of this disease with major practical implications for studies of disease mechanism and prevention .
Multiple sclerosis ( MS ) is a common inflammatory disease of the central nervous system characterized by myelin loss , axonal pathology , and progressive neurological dysfunction [1] . The aetiology of MS is unknown , however it is clear that genetic and environmental components are important [1] , [2] . The only genetic association with MS in Northern Europeans had been with extended MHC haplotypes , especially those containing HLA-DRB1*1501 [3] . The interleukin 7 receptor ( IL7RA ) , interleukin 2 receptor ( IL2RA ) , ecotropic viral integration site 5 ( EVI5 ) and kinesin family member 1B ( KIF1B ) genes have recently been shown to be additional MS susceptibility loci [4] , [5] , [6] , [7] . The largest of these , KIF1B , has a relatively small effect size ( odds ratio ( OR ) = 1 . 3 ) . The MHC ( OR = 5 . 4 ) is the key susceptibility locus in MS and other susceptibility genes identified to date appear to contribute little to overall risk [3] . The principal MHC class II haplotype that increases MS risk in individuals of Northern European descent is HLA- DQB1*0602-DQA1*0102 -DRB1*1501-DRB5*0101 [8] , although other HLA-DRB1 haplotypes have important influences on risk by epistatic interactions [9] , [10] , [11] , [12] . Intense linkage disequilibrium within the MHC has frustrated attempts at fine mapping and no precise susceptibility locus has been identified [9] , [13] . Twin studies have established that monozygotic ( MZ ) twin concordance is significantly greater than for dizygotics ( DZ ) . In the study by Willer and colleagues concordance was 25 . 3% and 5 . 4% respectively [14] . The observation that most MZ twin pairs are discordant for MS suggests environmental , stochastic factors or both but the most striking illustration of the importance of the environment in MS susceptibility is the 5-fold difference in MS risk between Tasmania and Queensland [15] . In the Northern Hemisphere , MS prevalence shows a north-south gradient , mirrored by a south-north gradient in the southern hemisphere ( reviewed by [16] ) . In accordance with the disease geography , sunlight , specifically through its role in generating active vitamin D , has been proposed as a key environmental factor for the disease [17] . Circumstantial evidence to support this comes from studies showing that MS patients are deficient in vitamin D [18] and that dietary vitamin intake reduces disease risk [19] . Additionally , a pooled analysis of over 40 , 000 patients from Canada , Great Britain , Denmark , and Sweden showed that fewer people with MS were born in November and more in May [20] , highlighting a risk factor that varies seasonally . Vitamin D is primarily known for its critical role in calcium homeostasis , however recent evidence has highlighted many actions on immune and central nervous system development and function [21] . These have contributed to the notion that this is how vitamin D affects MS risk , although direct links have not yet been identified . Vitamin D is a secosteroid hormone synthesized in the skin or ingested in the diet . Intake from dietary sources accounts for a much smaller proportion of total vitamin D , mainly owing to its rarity in foods [22] , [23] . During exposure to sunlight , ultraviolet B ( UVB ) radiation ( 290–315 nm ) is responsible for photolyzing 7-dehydrocholesterol , the precursor of vitamin D3 , to previtamin D3 which , in turn , rapidly spontaneously isomerizes to vitamin D3 [22] , [23] . Vitamin D3 is biologically inert and requires hydroxylation in the liver to 25-hydroxyvitamin D3 ( 25 ( OH ) D ) . Once formed , this major circulating form of vitamin D3 is further hydroxylated in the kidney to its active form , 1 , 25-dihydroxyvitamin D3 ( 1 , 25 ( OH ) 2D ) , by 25-hydroxyvitamin D-1α-hydroxylase ( 1-OHase ) . Recently it has been recognized that most tissues in the body ( including the brain , thymus and cells of the immune system ) also possess the 1-OHase enzyme . Thus numerous tissues in the body have the capacity to locally produce 1 , 25 ( OH ) 2D [22] , [23] . Most biological effects of 1 , 25-dihydroxyvitamin D3 or calcitriol , are mediated by the vitamin D receptor ( VDR ) . This receptor is a member of the steroid receptor super-family and influences the rate of transcription of vitamin D responsive genes by acting as a ligand activated transcription factor that binds to vitamin D response elements ( VDREs ) in gene promoters [21] . Early studies had provided evidence for an effect of vitamin D on HLA gene expression [24] , [25] , although no specific mechanism has been characterised . Here we examined the hypothesis of a direct interaction between vitamin D and MS associated MHC class II genes . Genetic variation characteristic of the most significant risk haplotypes for MS , those bearing HLA-DRB1*15 , includes a functional vitamin D response element ( VDRE ) in the proximal promoter region of HLA-DRB1 . This provides a mechanism linking the major environmental and genetic risk factors for MS .
Using the sequence for the HLA-DRB1*15 haplotype carried by the homozygous lymphoblastoid cell line PGF we scanned in silico for VDREs using Jaspar [26] with a profile score threshold of 80% . We analysed the entire genomic sequence of the HLA-DRB1 , HLA-DQA1 and HLA-DQB1 genes as well as 5 kb upstream of the transcriptional start sites of these genes to include promoter regions . VDREs exhibit a multitude of sequence variations , providing a spectrum of binding affinities for VDR , thus enabling these elements to respond to differing concentrations of VDR/1 , 25 ( OH ) 2D [22] . The analysis revealed only one potential VDRE located in the proximal promoter region immediately 5′ to the transcriptional start site of HLA-DRB1 ( Figure 1 ) . IL2RA and IL7RA were also searched in silico for potential VDR binding sequences; no putative VDREs were found . The occurrence and conservation of the putative VDRE element identified in the PGF sequence was examined in individuals with the HLA-DRB1*15 MS risk allele . The HLA-DRB1 promoter was resequenced in 322 HLA-DRB1*15 homozygous individuals , both MS affected and unaffected . An additional 168 individuals homozygous for other HLA-DRB1 alleles were also sequenced . The putative VDRE was present on all HLA-DRB1*15 bearing haplotypes with no variants found which disrupted the VDRE consensus sequence . In contrast , a number of nucleotide changes were found within the 15 base pairs of the VDRE on all non-HLA-DRB1*15 haplotypes . For example , nearly all ( 98% of 57 sequenced individuals ) of HLA-DRB1*04 , HLA-DRB1*07 and HLA-DRB1*09 haplotypes , all of which are non-MS associated alleles in the Canadian population [10] , carried the sequence GGGTGGAGAGGGGTCA . This sequence was predicted to function less effectively as a VDRE than the one on HLA-DRB1*15 bearing haplotypes according to Jaspar [26] . The modestly MS associated haplotype , HLA-DRB1*17 , differed from HLA-DRB1*15 at the VDRE in 50% of the individuals sequenced . The putative VDRE in the HLA-DRB1 promoter was investigated for ability to bind the vitamin D receptor in vitro using an electrophoretic mobility shift assay ( EMSA ) . Upon addition of recombinant VDR and retinoic acid receptor beta ( RXR , a co-regulator of VDR binding and transactivation [22] ) to a radiolabelled probe spanning the putative VDRE in the HLA-DRB1 promoter , two protein-DNA complexes on EMSA were observed ( Figure 2 , lane 2 ) . Both complexes were specifically competed with 10 to 100-fold molar excess of unlabelled VDRE probe ( Figure 2 , lanes 3–5 ) , while 10 to 100 fold molar excess of an unrelated probe containing an early growth response ( EGR ) factor binding site had no effect ( Figure 2 , lanes 6–7 ) . Finally , addition of a polyclonal antibody directed against VDR specifically retarded complex I , resulting in a supershift of the upper complex ( Figure 2 , lane 8 ) . This data showed the putative VDRE in the HLA-DRB1 promoter corresponding to the HLA-DRB1*15 haplotype could bind recombinant VDR/RXR with high specificity in vitro . When probes corresponding to the HLA-DRB1*04/07/09 variant VDRE were used , significantly lower affinity binding was found ( data not shown ) . Whether or not the VDR is recruited to the VDRE in the HLA-DRB1 gene promoter was examined ex vivo . Chromatin immunoprecipitation ( ChIP ) experiments were performed using lymphoblastoid cells bearing the HLA-DRB1*15 haplotype ( the PGF cell line ) which were either unstimulated or stimulated for 24 hours with 1 , 25-dihydroxyvitamin D3 and then cross-linked in the presence of formaldehyde . Immunoprecipitation was performed using antibodies against VDR . The VDR bound DNA fragments were then recovered after reversal of protein-DNA crosslinking and analysed by PCR using primers specific for the HLA-DRB1 promoter . A representative agarose gel is shown in Figure 3 . This revealed clear evidence of binding by VDR to the HLA-DRB1 promoter when compared to input chromatin and mock antibody controls for cells with the HLA-DRB1*15 haplotype , complementing the in vitro data from the EMSA experiments . The VDRE was then investigated to see if it modulated levels of gene expression in vitro . Reporter gene constructs were engineered in which −181 to +53 of the HLA-DRB1 gene sequence was placed upstream of a pGL3 luciferase reporter . pGL3_DRB1prom had the complete −181 to +53 sequence , pGL3_DRB1prom_hap1 had the same sequence as pGL3_DRB1prom but the VDRE replaced with the HLA-DRB1*04/07/09 VDRE and pGL3_DRB1prom_del had the 15 base pair VDRE sequence specifically deleted . These constructs were then transiently transfected into Raji B cells . A renilla luciferase reporter construct driven by the thymidine kinase promoter ( pRL_TK ) was co-transfected to normalise luciferase activity . pGL3_DRB1prom had significantly higher basal reporter gene activity than pGL3_DRB1prom_del ( P = 0 . 03 on paired t-test , two tailed ) . After stimulation with 1 , 25-dihydroxyvitamin D3 , there was a significant 1 . 6 fold increase in luciferase activity with pGL3_DRB1prom ( P = 0 . 002 ) , but no significant change with pGL3_DRB1prom_del ( P = 0 . 12 ) , nor pGL3_DRB1prom_hap1 ( P = 0 . 58 ) ( Figure 4 ) . To investigate any effect of vitamin D on the cell surface expression of HLA-DRB1 , the HLA-DRB1*15 homozygous lymphoblastoid cell line PGF and the HLA-DRB1*07 homozygous lymphoblastoid DBB cell line were stained with anti-HLA-DRB1 antibody . PGF cells constitutively expressed HLA-DRB1 at higher levels then DBB ( average geometric mean fluorescence intensity ( MFI ) PGF = 97 . 1 , DBB = 42 . 8 , P = 0 . 0002 ) . Upon addition of 1 , 25-dihydroxyvitamin D3 , there was a 1 . 3 fold increase in the expression of HLA-DRB1 in PGF cells ( P = 0 . 031 on paired t-test , two tailed ) but no significant difference in the expression of HLA-DRB1 in DBB cells ( P = 0 . 10 ) .
While the role of the environment is clearly important in determining MS risk , the relevant underlying mechanism ( s ) have remained elusive and there has been no experimental support for a direct environment-gene interaction . Although differences in Epstein-Barr virus infection are seen when MS patients are compared to controls , extensive searches for specific viral infections have failed to confirm direct involvement . [2] . Where appropriate data is available , the amount of winter sunlight parallels the range of MS prevalence , and high sunlight exposure is associated with low disease prevalence [2] . The effects of migration between high and low risk geographic regions have been examined in several populations ( e . g . UK immigrants to South Africa , or Asian and Caribbean immigrants to the UK ) . These studies show that MS risk is influenced by the migrant's country of origin [27] . Despite the limits of small sample sizes , a ‘critical age’ has been hypothesized: immigrants who migrate before adolescence acquire the risk of their new country , while those who migrate after retain the risk of their home country . Dietary difference for vitamin D intake ( oily fish consumption ) plausibly explains the striking exception to MS latitudinal risk in Norway [2] . As familial aggregation is genetically determined [28] , environmental factors thus appear to be operative at a broad population level , perhaps acting at a young age [27] and/or during gestation [20] . A good candidate for an environmental factor that influences MS disease risk is vitamin D . We approached the candidacy of vitamin D by searching first for vitamin D response elements within the MHC class II region . Specifically we investigated the major candidate genes in the disease associated locus , HLA-DRB1 , HLA-DQA1 and HLA-DQB1 and identified a consensus binding site for VDR next to the HLA-DRB1 gene . This was the only VDRE we found and strikingly it shows haplotype-specific differences , being highly conserved in the major MS associated haplotype HLA-DRB1*15 dominant in Northern European populations , but not conserved among non-MS associated haplotypes . This was itself circumstantial evidence supporting a vitamin D role in the functional characteristics of this haplotype . The identified VDRE lies close to the highly conserved MHC class II specific regulatory SXY module . This module comprises S , X and Y regulatory elements important for constitutive , and indirectly for IFN-γ-induced , expression of HLA class II genes co-ordinated by the MHC class II transactivator MHC2TA [29] . The VDRE was highly conserved on HLA-DRB1*15 haplotypes ( no mutations on over 600 chromosomes ) suggesting a selective pressure to maintain this response element for the HLA-DRB1*15 allele . Variants were found to some extent on all other non HLA-DRB1*15 haplotypes . The results may additionally/alternatively reflect the ancestral origin of the HLA-DRB1*15 ( DR51 ) haplotype [30] which displays the strongest linkage disequilibrium among the MHC class II haplotypes [31] . We note the association between this haplotype and MS risk is characteristic of Northern European populations , the ones most vulnerable to vitamin D deficiency [2] . EMSA experiments using recombinant proteins demonstrated that in vitro VDR can bind specifically to the putative VDRE in the proximal HLA-DRB1 promoter found on the HLA-DRB1*15 haplotype . ChIP data showed specific enrichment of the region spanning the VDRE in VDR immunoprecipitated samples relative to input and mock antibody controls , demonstrating that the vitamin D receptor was recruited to this haplotype in this ex vivo model system . Finally , transient transfection and flow cytometric assays established that the VDRE present in the HLA-DRB1 promoter can influence gene expression and imparts 1 , 25-dihydroxyvitamin D3 sensitivity to HLA-DRB1*15 . The variant VDRE present on other , non-MS associated HLA-DRB1 haplotypes was not responsive to 1 , 25-dihydroxyvitamin D3 . A T cell repertoire with millions of specificities provides surveillance against a multitude of foreign pathogens [32] . An inherent danger in recognizing so many foreign proteins is the potential to respond to self-proteins . To circumvent this problem T cells are scrutinised for self-reactivity as they mature in the thymus with deletion of those posing the greatest threat ( central deletion ) [32] . One constraint on central deletion is the requirement for the relevant autoantigen to be present in the thymus . Whether or not these are expressed as proteins at levels sufficient to induce T cell deletion is not clear . Given the results of this study , variable expression of HLA-DRB1 could affect central deletion of autoreactive T cells . It is plausible that a lack of vitamin D in utero or early childhood can affect the expression of HLA-DRB1 in the thymus , and impacting on central deletion . For MS , in HLA-DRB1*15 bearing individuals , a lack of vitamin D during early life could allow auto reactive T cells to escape thymic deletion and thus increase autoimmune disease risk . Indeed it has been shown that antigen presentation in the thymus of VDR knock-out mice is impaired [33] . However the mechanism for a HLA- vitamin D interaction remains unclear as is the timing and tissue in which such interactions might occur . A major selective pressure on skin pigmentation is thought to have been vitamin D deficiency with progressively lighter skin pigmentation at increasing distance from the equator related to variation in intensity of ultraviolet radiation with latitude [34] . The presence of a VDRE specific to HLA-DRB1*15- bearing haplotypes , present at high allele frequencies among Northern Europeans , suggests a possible role for vitamin D in selection at this locus . The intriguing possibility that vitamin D responsiveness rather than any antigen-specificity determines the increased MS risk of the HLA-DRB1*15 haplotype warrants consideration and can be tested in the infrequent haplotypes bearing the VDRE on other non-HLA-DRB1*15 haplotypes . In summary , we have identified and functionally characterised a vitamin D response element ( VDRE ) in the HLA-DRB1 promoter region . These studies imply direct interactions between HLA-DRB1 , the main susceptibility locus for MS , and vitamin D , a strong candidate for mediating the environmental effect . This study provides more direct support for the already strong epidemiological evidence implicating sunlight and vitamin D in the determination of MS risk . Given that a high frequency of vitamin D insufficiency in the general population has been observed [35] , our data support the case for supplementation during critical time periods to reduce the prevalence of this devastating disease .
All participants in the study were ascertained through the ongoing Canadian Collaborative Project on the Genetic Susceptibility to MS ( CCPGSMS ) [36] . Subject ascertainment , genotyping and sequencing has been previously described [9] , [10] , [37] . Each participating clinic in the CCPGSMS obtained ethical approval from the relevant institutional review board , and the entire project was reviewed and approved by the University of British Columbia and the University of Western Ontario . EMSAs were performed as previously described [38] . The VDRE probe comprised of the annealed sense and antisense strands of the nucleotide sequence agctGTGGGTGGAGGGGTTCATAG , the EGR probe agctAAATCCCCGCCCCCGCGATGGA and the VDRE variant probe agctGTGGGTGGAGAGGGGTCATAG . Full length recombinant purified VDR and recombinant purified RXR beta were purchased from Invitrogen , and polyclonal VDR antibody from Affinity Bioreagents . Radioactivity was quantitated with the Packard Cyclone phosphorimager , and analyzed with Optiquant ( Perkin Elmer Life Sciences ) . Values were compared using the Chi square test . The lymphoblastoid cell line PGF was cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum , 0 . 2 mM L-glutamine at 37°C in 5% humidified CO2 . 60×106 cells were harvested unstimulated or after stimulation with 0 . 1 uM calcitriol ( Sigma ) . Cells were crosslinked using a 1% formaldehyde buffer for 15 minutes at room temperature , quenched with glycine and chromatin prepared as previously described [39] . Chromatin was sheared by sonication in the presence of 212–300 microns glass beads ( Sigma ) at 4°C using a double step microtip attached to a Branson 450 Sonifier with coupler ( Branson ) in 30 second bursts ( six pulses at 40% ) with the samples cooled on ice for 1 minute between pulses . Sonicated chromatin was then processed and subject to immunoprecipitation as previously described [39] using magnetic ‘Dynabeads M-280’ ( Dynal ) precoated with anti rabbit IgG to which the primary antibody VDR was bound ( Affinity Bioreagents ) . We followed the buffer used for immunoprecipitation and subsequent washes as described [40] . Following reversal of crosslinks , RNase A and Proteinase K digestion , DNA was extracted using phenol-chloroform and amplified by PCR with separation on a 2 . 0% agarose gel . The primers used for PCR were: forward- GCAACTGGTTCAAACCTTCC and reverse- GTCCCCAGACAAAGCCAGT . Cycling conditions were: 95°C for 10 minutes; a touchdown of 14 cycles ( 95°C for 30 seconds; 61°C with −0 . 5°C per cycle , for 30 seconds; 72°C for 30 seconds ) ; 35 cycles of 95°C for 30 seconds , 53 . 5°C for 30 seconds , 72°C for 30 seconds; 72°C for 7 minutes . The plasmids were constructed by inserting the promoter region ( −181 to +53 ) of the human HLA-DRB1 gene ( pGL3_DRB1prom with the VDRE sequence ( chr6:32 , 665 , 500–32 , 665 , 760 ) , pGL3_DRB1prom_del with the VDRE sequence deleted ( chr6:32 , 665 , 500–32 , 665 , 559 combined with chr6:32 , 665 , 575–32 , 665 , 760 ) ) into the pGL3 reporter plasmid . Two independent plasmid preparations were used in transient transfection experiments for each construct . Raji B cells were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum , 0 . 2 mM L-glutamine at 37°C in 5% humidified CO2 . Lipofectamine-LTX and PLUS reagent ( Invitrogen ) were used for transient transfection of expression constructs , following the manufacturer's protocol . pRL_TK was co-transfected to normalize for transfection efficiency . When indicated , cells were stimulated with 0 . 1 uM calcitriol ( Sigma ) for 24 hours . Cells were harvested after 24 hours and lysed in 500 ul of 1× lysis buffer ( Promega ) and analyzed using the Dual-Luciferase reporter assay kit ( Promega ) and a Turner luminometer model 20 ( Promega ) following the manufacturer's protocol . Paired t-tests were used to compare expression values . Each transfection was carried out 12 times in total . The lymphoblastoid cell lines PGF ( International Histocompatibility Workshop number IHW09318 ) and DBB ( IHW09052 ) were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum , 0 . 2 mM L-glutamine at 37°C in 5% humidified CO2 . 1×106 cells were harvested unstimulated or 24 hours after stimulation with 0 . 1 uM calcitriol ( Sigma ) in three biological replicates . Cells were stained with either a FITC conjugated monoclonal anti-human HLA-DR antibody ( Sigma , F1902 ) or a FITC conjugated isotype control antibody ( Sigma , F6522 ) for 30 minutes at room temp , then washed with 2% BSA in PBS and re-suspended in 1 mL of 2% paraformaldehyde . Cells were analysed using CyAn flow cytometer ( Dako ) .
|
Multiple Sclerosis ( MS ) is a complex neurological disease with a strong genetic component . The Major Histocompatibility Complex ( MHC ) on chromosome 6 exerts the strongest genetic effect on disease risk . A region at or near the HLA-DRB1 locus in the MHC influences the risk of MS . HLA-DRB1 has over 400 different alleles . The dominant haplotype of Northern Europe , marked by the presence of DRB1*1501 , increases risk of MS by 3-fold . The environment also plays a key role in MS . The most striking illustration of this is the geographical distribution of the disease in populations matched for ethnicity . This has led to the proposal that sunshine , and in particular , vitamin D , is an environmental factor influencing the risk of MS . Circumstantial evidence supporting this comes from studies showing the involvement of vitamin D in immune and nervous system function . The current investigation sought to uncover any relationship between vitamin D and HLA-DRB1 . It was found that vitamin D specifically interacts with HLA-DRB1*1501 to influence its expression . This study therefore provides more direct support for the already strong epidemiological evidence implicating sunlight and vitamin D in the determination of MS risk , and implies that vitamin D supplementation at critical time periods may be key to disease prevention .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"public",
"health",
"and",
"epidemiology/epidemiology",
"immunology/genetics",
"of",
"the",
"immune",
"system",
"neurological",
"disorders/multiple",
"sclerosis",
"and",
"related",
"disorders",
"immunology/autoimmunity"
] |
2009
|
Expression of the Multiple Sclerosis-Associated MHC Class II Allele HLA-DRB1*1501 Is Regulated by Vitamin D
|
The treatment of Leishmaniasis caused by Leishmania ( Viannia ) guyanensis is based on a weak strength of evidence from very few clinical trials and some case series reports . Current treatment guidelines recommend pentamidine isethionate or meglumine antimoniate ( Glucantime ) as the first-line choices . Both are parenteral drugs with a low therapeutic indexes leading to a high risk of undesired effects . Imidazole derivatives interfere with the production of leishmanial ergosterol , an essential component of their membrane structure . One drug that has been studied in different clinical presentations of Leishmania is fluconazole , a hydrophilic bis-triazole , which is easily absorbed through the oral route with a low toxicity profile and is considered safe for children . This drug is readily available in poor countries with a reasonable cost making it a potential option for treating leishmaniasis . An adaptive nonrandomized clinical trial with sequential groups with dose escalation of oral fluconazole was designed to treat adult men with localized cutaneous leishmaniasis ( LCL ) in Manaus , Brazil . Eligible participants were patients with LCL with confirmed Leishmania guyanensis infection . Twenty adult male patients were treated with 450 mg of fluconazole daily for 30 days . One patient ( 5% ) was cured within 30 days of treatment . Of the 19 failures ( 95% ) , 13 developed a worsening of ulcers and six evolved lymphatic spreading of the disease . Planned dose escalation was suspended after the disappointing failure rate during the first stage of the trial . Oral fluconazole , at the dose of 450mg per day , was not efficacious against LCL caused by Leishmania guyanensis in adult men . Brazilian Clinical Trial Registration ( ReBec ) —RBR-8w292w; UTN number—1158-2421
Cutaneous leishmaniasis ( CL ) is one of the 17 neglected diseases , with 350 million people at risk in 98 endemic countries [1] . Latin America is a highly endemic region with a documented 30% increase in the number of reported cases during a period of ten years ( 2001–2011 ) [2] . Brazil together with eight other countries reports 90% of all registered cases of CL [1] . The incidence of CL in the Brazilian Amazon region surpasses the Brazilian national average ( 15 . 3 cases/100 . 000 inhabitants ) , reaching in some regions the incidence of 30 cases/100 , 000 inhabitants [3] . American tegumentary leishmaniasis ( ATL ) , especially in the Amazon region , has a peculiar epidemiological profile characterized by higher intra- and inter-specific variations and heterogeneity of transmission cycles , reservoir hosts and sand fly vectors , with sympatric circulation of various Leishmania species . These characteristics altogether lead to diverse clinical presentations and distinct clinical responses to treatment [4] . Leishmania guyanensis is the most common cause of human leishmaniasis in the Guianan Ecoregion Complex ( GEC ) a region that covers Guiana , Suriname , French Guiana , the southern portion of Venezuela and the northern portion of the Amazon Basin [5 , 6] . Cutaneous leishmaniasis is the most common clinical presentation of the infection caused by Leishmania guyanensis [7] . Lymphatic involvement , manifested as adenomegaly or lymphangitis , is the second most common presentation of the disease affecting 60% of the cases [8] . Disseminated cutaneous leishmaniasis and mucosal disease are less common presentations of leishmaniasis caused by this Leishmania species [9 , 10] . The current treatment regimens for CL caused by Leishmania guyanensis follow a weak strength of recommendation based on a low-quality evidence [11] . Intramuscular administration of 3 mg/kg pentamidine isethionate every other day for up to four injections is considered the treatment of choice [12] . Meglumine antimoniate ( Glucantime ) 20mg Sv/Kg/day for 20 days is the recommended treatment in Brazil with a cure ratebetween 53% and 70% [13 , 14] . Miltefosine has been used in some studies with a cure rate between 54% and 72% [15 , 16] , although that drug is not available in South America . Amphotericin B is the drug choice in severe cases . The present treatment options are parenteral drugs with low cure rates and low therapeutic indexes leading to a high risk for undesired effects . This scenario leads us to research a new therapeutic option . One drug that has been studied in different clinical presentations of Leishmania is fluconazole . This agent is a hydrophilic bis-triazole that is easily absorbed via the oral route and interferes with the production of leishmanial ergosterol , an essential component of the membrane structure [17] . Fluconazole , has a low toxicity profile , it accumulates rapidly and extensively in the skin , and it is readily available , with a reasonable cost . Fluconazole is also considered safe for children [18 , 19] . This adaptive phase II trial evaluated the efficacy of fluconazole in the treatment of cutaneous leishmaniasis caused by L . guyanensis .
From December 2014 through February 2016 , twenty-eight subjects with parasitological confirmed diagnosis of LCL were recruited at the Tropical Medicine Foundation Dr . Heitor Vieira Dourado , in the state of Amazonas , North Brazil , an endemic area of L . guyanensis infection . The flow diagram of participants through the different study phases is described in Fig 1 . Localized cutaneous leishmaniasis was defined as the presence of up to five ulcerous lesions without lymphatic or mucousal disease , with amastigotes visualized in direct examination of Giemsa-stained smears of a dermal scrapping taken from the ulcerated border of at least one lesion . Inclusion criteria were as follows: ( 1 ) diagnosis of LCL based on case definition , ( 2 ) illness duration <3 months , ( 3 ) male sex with age of at least 18 years , ( 4 ) 1 to 5 ulcerated lesions , and ( 5 ) no previous treatment for leishmaniasis . Exclusion criteria were as follows: ( 1 ) Leishmania species could not be identified; ( 2 ) infection caused by species other than L . guyanensis; ( 3 ) any uncontrolled active infectious or severe disease , and ( 4 ) an allergy to fluconazole . Complete blood cell count , tests for the levels of aspartate and alanine aminotransferase , amylase , lipase , urea , creatinine and glucose , and an electrocardiogram were performed in all participants before therapy and at 30 , 60 , 90 and 180 days after treatment . All patients were subjected to a rapid human immunodeficiency virus test and serology for hepatitis B and C . All the ulcers were measured and photographed . A biopsy of each ulcer was performed and the material was used for a parasite culture and histopathology . Leishmania species were identified as described by Marfurt et al [20] . Scheduled patient visits were made 30 , 60 , 90 and 180 days after beginning the treatment . If a patient did not return to follow-up at the specified time , visits were conducted in the patient’s home on the same day or within 7 days of the missed appointment . Patients’ ulcers were measured with a flexible ruler at the initial visit and at each follow-up visit . Standardized digital photographs of the patients’ lesions were obtained at the same time points . Patients were monitored for adverse events ( AEs ) and treatment adherence . Patients returned the blister packs of fluconazole to verify compliance . Clinical and laboratory AEs were graded according to the Common Terminology Criteria for Adverse Events of the National Cancer Institute [21] . Outcome measures followed the protocols published by Olliaro et al [22] . The primary endpoint was a definitive cure six months after the end of treatment . A definitive cure was defined as the complete epithelialization of all lesions without raised borders , infiltrations , inflammation or crusts . The secondary endpoint included an initial cure defined as complete epithelialization of ulcers two months after the end of treatment . If an initial cure was not attained it was considered a therapeutic failure . Any interruption of the treatment was also considered a therapeutic failure . All patients included in the therapeutic failure group received a rescue therapy of meglumine antimoniate , 20mg Sbv/Kg/day for 20 days . Fluconazole 150 mg capsules conditioned in blisters packs containing 10 capsules were self-administered orally for 30 days . The first dosing schedule involved 450 mg of fluconazole administered once daily , and the second dosing schedule involved 900 mg of fluconazole administered as two daily doses of 450 mg . Both schedules lasted 30 days . An adaptive phase II trial was adopted based on the successful use of fluconazole in treating L . ( Viannia ) braziliensis in a study published by Sousa et . al [23] . The adaptive trial followed FDA recommendations [24] . The sample size of 30 patients in the first step of the study was calculated considering an estimated cure rate of 60% , a precision of +/- 20% and an alpha error of 5% . The patient was instructed to take three capsules ( 450mg total dose ) orally once daily in the morning for 30 days . If 18 patients had reached the primary endpoint with this regimen , then the second step of the study would have begun . The second step was designed based on the assumption of an improvement in magnitude of response of at least 10% after doubling the dose of fluconazole . Considering a cure rate of 70% , a precision of +/- 10% and an alpha error of 5% , a sample size of 73 patients was calculated . Patients in the second step would have received three 150 mg fluconazol in the morning and three 150 mg fluconazole in the afternoon ( 900 mg total dose ) for 30 days . All statistical analyses were performed with SPSS 21 . 0 software for Windows . This trial was conducted according to the Declaration of Helsinki . Before they were enrolled in the study , written informed consent was obtained from all patients . The study was approved by the Ethics Committee of the Tropical Medicine Foundation Dr Heitor Vieira Dourado , Brazil - registration number 26118613 . 4 . 0000 . 0005 . This clinical trial was registered in ReBEC ( Brazilian Registry of Clinical Trials ) with the identifier RBR-8w292w and is available from http://www . ensaiosclinicos . gov . br/rg/view/2668/ UTN number–1158–2421
Twenty-eight adult male patients fulfilled the inclusion criteria and were accepted into the study . Their median age was 38 . 3 years old ( range 18–56 ) . Of those , eight ( 28 . 6% ) were excluded . In three cases the causative species was not Leishmania guyanensis , and in the other five , the parasite species could not be identified . Twenty adult male patients ( 71 . 4% ) remained in the study for further analysis . Those 20 patients presented 40 lesions , all with less than 3 months of duration . Fifty-five percent ( n = 11 ) presented with one lesion , 15% ( n = 3 ) with two lesions , 15% ( n = 3 ) with three lesions , 5% ( n = 1 ) with four lesions and 10% ( n = 2 ) with five lesions . Concerning the distribution of the lesions , they were predominantly on exposed areas of the body , with 37 . 5% ( n = 15 ) located on the lower limbs , 37 . 5% ( n = 15 ) located on the upper limbs and 15% ( n = 6 ) located on the face . Ten percent of the lesions ( n = 4 ) were located on the trunk . The median diameter of all 40 ulcers on the first day of treatment was 1 . 70cm ( range 0 . 2 cm– 4 cm ) . During the trial , five participants with 17 total ulcers were excluded before the end of 30 days of treatment . At the end of treatment ( day 30 ) , 23 lesions on 15 patients remained active ulcers with a median diameter of 2 . 80 cm ( range 0 . 4 cm– 5 cm ) . Twelve lesions on six patients remained active ulcers at day 60 ( 30 days after the end of treatment ) , with a median diameter of 2 . 37 cm ( range 0 . 3 cm– 4 . 5 cm ) . The clinical evolution of the ulcers is represented in Fig 2 . A definitive cure was documented in 5% ( n = 1 ) of the cases , as shown in S1 Fig . The remaining 19 patients were considered treatment failures . Five patients asked to change medication due to ulcer enlargement . In eleven cases ( 55% ) there was worsening of the ulcers with marked inflammatory signs ( Fig 3 ) . In six patients ( 30% ) lymphatic spread of the disease was also noted ( Fig 3 ) . In those 19 patients , fluconazole treatment was stopped and rescue therapy was instituted with complete resolution of the ulcers . The drug was well tolerated with mild self-limited systemic adverse events in five patients ( 27 . 78% ) as shown in Fig 4 .
The first report of the activity of an azole against a species of the genus Leishmania came from an in vitro test of CIBA 32 , 644-Ba in 1965 and another of 2-amino-5- ( 1-methyl-5-nitro-2-imidazolyl ) 1-3-thiadiazole in1968 [25 , 26] . At the end of the 1960s and in the 1970s , in Brazil , the clinical efficacy of 1- ( 5-nitro-2-tiazolil ) 2-imidazolidinone and niridazole in cutaneous and mucosal leishmaniasis caused by L . ( Viannia ) braziliensis was reported , in case series , with some clinical response , at the cost of serious neurologic adverse events [27 , 28] . The interest in using azoles in the treatment of leishmaniasis was revived after the report published by Berman indicating the activity of ketoconazole against leishmanial species in macrophage culture [29] . The efficacy of the azoles in treating ATL is mainly based on case series or small trials with heterogeneous results . The cure rate of ketoconazole 400mg bid was as follows: one out of six patients with ATL caused by L . guyanensis [30] , three out of three patients with ATL caused by L . braziliensis [31] and 16 out of 22 patients with ATL by L . panamensis [32] . In two studies , itraconazole cured six cases out of ten [33] in one study and three out of 13 patients in the other [34] . The first clinical use of fluconazole in leishmaniasis was against kala-azar with 0% definite cure . Some patients had early apparent cures with later relapsing [35] . In 2002 , Alrajhi et al . published a randomized , placebo-controlled trial and concluded that a six-week course of 200 mg fluconazole daily was safe and useful to treat CL caused by L . major [36] . Afterwards , this drug became an alternative for the treatment of Old World cutaneous leishmaniasis . A trial published by Emad et al . evidenced that 400mg of fluconazole daily was more efficacious in infections caused by L . major when compared to 200mg daily , with six-week cure rates of 81% versus 48 . 3% respectively [37] . In Brazil , a case series was conducted where 28 patients with confirmed leishmaniasis caused by L . ( V . ) braziliensis , who refused or could not use antimonials , received oral fluconazole for 20 days . Eight patients received 5mg/Kg/day with a cure rate of 75% , 14 patients received 6 . 5mg/Kg/day with a cure rate of 92 . 8% and six patients received 8mg/Kg/day with a cure rate of 100% [23] . The authors concluded that there was a higher efficacy at higher doses of fluconazole . Treatment of leishmaniasis caused by L . guyanensis follows a weak strength of recommendation based on a low-quality of evidence [11] . Eight published trials analyzed the cure rate in cases of ATL caused by this species , with results varying from 53 . 6% to 91 . 7% ( S1 Table ) . Considering the lack of an optimal treatment for ATL caused by L . guyanensis , fluconazole seemed to be a promising drug alternative for treating this disease . This assumption was based on the mechanism of action of the drug in the protozoan ergosterol metabolism and on the clinical efficacy evidenced in clinical trials of Old World leishmaniasis and the clinical response against L . braziliensis [23] . In an adaptive clinical trial design it is possible to evaluate the desired outcomes exposing fewer patients without losing the quality of the evidence [38] . The sequential groups with scaled doses allow timely suspension of the research after initial failures with a smaller dose , minimizing unnecessary drug exposure . During the execution of the study , the first clinical trial evaluating fluconazole in ATL caused by L . braziliensis was published . The intention-to-treat analysis two months after treatment showed cure rates of 22 . 2% ( 6 out of 27 ) in the fluconazole group and 53 . 8% ( 14 out of 26 ) in the Glucantime group . The per protocol results were the same at six months after the end of treatment . The AEs were similar in both groups [39] . In this study , the cure observed with 450 mg of fluconazole daily against L . guyanensis was 5% ( 1 out of 20 ) . Five patients asked to stop taking fluconazole and to receive the rescue therapy . One patient who had clinical failure refused to receive any parenteral medication , and after 120 days he presented no signs of the disease . It is not clear if this outcome was a spontaneous evolution to a cure or a delayed response to the fluconazole treatment . The remaining 11 patients presented a peculiar clinical outcome . The ulcers , after a period of ten days , began to show remarkable inflammatory signs associated with intense pain , increased size and , in some cases the lymphatic spread of the disease ( Fig 3 ) . The early enlargement of the ulcers during treatment with meglumine antimoniate was previously reported , but the intensity of inflammatory signs and the concomitant lymphatic involvement constitutes a novel finding never reported in patients with leishmaniasis exposed to fluconazole [40] . The mechanism of this inflammatory phenomenon deserves more investigation , considering that this high failure rate may be justified by an unrecognized immunological mediated effect associated to fluconazole . All 16 patients that received Glucantime as rescue therapy were cured after 30 days . One patient moved from Manaus and the researchers lost contact . Fluconazole was well tolerated systemically , with five cases of mild AEs . In conclusion , fluconazole , at the dose of 450mg per day , is not efficacious against leishmaniasis in adult men infected by L . guyanensis . The clinical worsening during fluconazol exposure deserves more attention and should be evaluated in future studies involving fluconazole or other azole therapies .
|
The main agent causing localized cutaneous leishmaniasis in the Guianan Ecoregion Complex , a region that covers Guiana , Suriname , French Guiana , the southern portion of Venezuela and the northern portion of the Amazon Basin , is Leishmania ( Viannia ) guyanensis . The current treatment regimens for treating this parasite follow a weak strength of recommendation based on a low-quality evidence . All the drugs used are parenteral drugs with low therapeutic indexes and inadequate cure rates . Azoles are drugs that interfere with the production of leishmanial ergosterol and have been studied in other Leishmania species with promising results . Fluconazole is a hydrophilic bis-triazole that accumulates rapidly and extensively in the skin with a low toxicity profile . It is considered safe for children . Oral fluconazole is readily available and has a reasonable cost . In this work , we estimated the efficacy of fluconazole in treating localized cutaneous leishmaniasis caused by L . guyanensis in Manaus . The study had to be stopped due to the disappointing results , with only a 5% cure rate . Among the failures , 55% developed a marked worsening of their ulcers and 55% of those patients had lymphatic dissemination of the infection . In conclusion , fluconazole at the dose of 450mg per day is not efficacious in treating localized cutaneous leishmaniasis caused by L . guyanensis .
|
[
"Abstract",
"Introduction",
"Methods",
"and",
"material",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"clinical",
"research",
"design",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"organic",
"compounds",
"research",
"design",
"clinical",
"medicine",
"protozoans",
"signs",
"and",
"symptoms",
"leishmania",
"ulcers",
"neglected",
"tropical",
"diseases",
"pharmacology",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"heterocyclic",
"compounds",
"south",
"america",
"azoles",
"lesions",
"chemistry",
"adverse",
"events",
"protozoan",
"infections",
"brazil",
"people",
"and",
"places",
"eukaryota",
"diagnostic",
"medicine",
"drug",
"research",
"and",
"development",
"organic",
"chemistry",
"clinical",
"trials",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] |
2018
|
Failure of fluconazole in treating cutaneous leishmaniasis caused by Leishmania guyanensis in the Brazilian Amazon: An open, nonrandomized phase 2 trial
|
The recently discovered human Merkel cell polyomavirus ( MCPyV or MCV ) causes the aggressive Merkel cell carcinoma ( MCC ) in the skin of immunocompromised individuals . Conflicting reports suggest that cellular glycans containing sialic acid ( Neu5Ac ) may play a role in MCPyV infectious entry . To address this question , we solved X-ray structures of the MCPyV major capsid protein VP1 both alone and in complex with several sialylated oligosaccharides . A shallow binding site on the apical surface of the VP1 capsomer recognizes the disaccharide Neu5Ac-α2 , 3-Gal through a complex network of interactions . MCPyV engages Neu5Ac in an orientation and with contacts that differ markedly from those observed in other polyomavirus complexes with sialylated receptors . Mutations in the Neu5Ac binding site abolish MCPyV infection , highlighting the relevance of the Neu5Ac interaction for MCPyV entry . Our study thus provides a powerful platform for the development of MCPyV-specific vaccines and antivirals . Interestingly , engagement of sialic acid does not interfere with initial attachment of MCPyV to cells , consistent with a previous proposal that attachment is mediated by a class of non-sialylated carbohydrates called glycosaminoglycans . Our results therefore suggest a model in which sialylated glycans serve as secondary , post-attachment co-receptors during MCPyV infectious entry . Since cell-surface glycans typically serve as primary attachment receptors for many viruses , we identify here a new role for glycans in mediating , and perhaps even modulating , post-attachment entry processes .
The human Merkel cell polyomavirus ( MCPyV or MCV ) was discovered in 2008 and found to be clonally integrated into Merkel cell carcinomas ( MCCs ) , establishing it as the first human oncovirus from the polyomavirus family [1] . MCPyV infection is common , with 50–80% of adults being seropositive [2] . It establishes persistent asymptomatic infections in the skin of healthy individuals , many of whom chronically shed virions [3] . In immunocompromised individuals , MCCs arise from the malignant transformation of mechanoreceptor Merkel cells in the skin by the transforming antigens of MCPyV [4] , [5] . MCC is lethal , with an overall 5-year survival of MCC of only 50% , and its incidence has increased to 1 , 500 new cases per year in the USA alone [6] , [7] . There are no vaccines or antivirals against MCPyV . Polyomaviruses are non-enveloped , double-stranded DNA viruses that infect mammals and birds . There are currently nine human polyomaviruses , seven of which have been identified in the last five years [3] , [8] , [9] , [10] , [11] , [12] . Similar to MCPyV , the human BK and JC Polyomaviruses ( BKPyV and JCPyV ) establish persistent asymptomatic infections but cause severe disease in immunosuppressed individuals [2] . Although polyomaviruses such as Simian Virus 40 ( SV40 ) and Murine Polyomavirus ( mPyV ) can transform cells in culture or cause tumors in animal models , MCPyV is the first virus in the family that has been clearly implicated as a causal agent underlying a human cancer [1] , [4] , [5] , [13] . Polyomavirus infectious entry is initiated by the major capsid protein VP1 , which attaches to cellular receptors to promote internalization and transport of the viral genome into the nucleus for replication [14] . MCPyV uses sulfated carbohydrates termed glycosaminoglycans ( GAGs ) as attachment receptors . This contrasts with better-studied polyomaviruses , such as murine polyomavirus ( mPyV ) , SV40 , BKPyV and JCPyV , which use carbohydrates containing sialic acid for cell attachment and internalization [15] , [16] , [17] . Sialic acids have nevertheless been implicated in MCPyV infection as cell lines lacking sialylated glycans are resistant to transduction with an MCPyV reporter virus [18] . MCPyV VP1 has also been shown to interact in vitro with the ganglioside GT1b , which carries three sialic acids [19] . However , it was not understood in which way sialic acids are involved in MCPyV infection , nor whether a direct interaction with sialic acid is required for productive infection . Sialic acids cap N- and O-linked glycoproteins as well as glycolipids and are found on all eukaryotic cell surfaces . The most common sialic acid in humans is N-acetyl neuraminic acid ( Neu5Ac ) [20] , [21] . Structural studies of VP1-receptor complexes from other polyomaviruses have elucidated their interactions with different sialylated oligosaccharides [17] , [22] , [23] . However , a putative MCPyV sialic acid binding site must differ from previously characterized ones as MCPyV lacks conserved residues that engage sialic acids in other polyomaviruses . Thus , the structural basis of MCPyV's requirement for sialylated glycans remains unknown . In this study , we present crystal structures of MCPyV VP1 in complex with sialylated oligosaccharides . Analysis of the observed interactions in solution using NMR spectroscopy allows us to identify a linear Neu5Ac-α2 , 3-Gal disaccharide as the motif recognized by MCPyV . Based on the structural information , we conduct mutagenesis experiments that directly establish the functional relevance of the interaction with sialic acid for MCPyV infection . Our results therefore illuminate a crucial post-attachment interaction event of MCPyV , providing a foundation for the development of antiviral strategies .
We solved the crystal structure of unassembled MCPyV VP1 pentamers at 2 . 1 Å resolution ( Table 1 ) . The crystallized VP1 construct was truncated at the C-terminus to prevent VP1 assembly into capsids , and at the N-terminus to remove potentially disordered residues that inhibit crystallization . However , the construct contained the entire pentameric core of VP1 . Similar truncations did not have an effect on the receptor binding properties of other polyomavirus VP1 proteins [17] , [22] , [23] , [24] . MCPyV VP1 is a symmetric ring-shaped homopentamer with the five VP1 monomers arranged around a central five-fold axis ( Fig . 1A ) . Each monomer is composed of two antiparallel β-sheets , which together form a β-sandwich with jelly-roll topology . With β-strands named alphabetically from the N-terminus , the two sheets consist of strands B , I , D , G , and C , H , E , F , respectively . The β-strands are linked by extensive loops that cover the top and sides of the pentamer . The apical loops , which make up the top surface of the pentamer and thus the outer surface of the virus , are the most variable parts among VP1 sequences from different polyomaviruses , creating unique interaction surfaces . We next determined high-resolution structures of MCPyV VP1 in complex with three different sialylated oligosaccharides derived from the in vitro binding partner GT1b ( Fig . 1 ) . 3′-Sialyllactosamine ( 3SLN ) is a linear compound containing a single α2 , 3-linked Neu5Ac residue ( Fig . 1D ) . Disialyllactose ( DSL ) is also linear , carrying a second , α2 , 8-linked Neu5Ac attached to the one present in 3SLN . Both 3SLN and DSL are similar to the carbohydrate portions of gangliosides ( GM3 and GD3 , respectively ) , but they also can be found capping the carbohydrate parts of glycoproteins . GD1a , the oligosaccharide portion of the GD1a ganglioside , is a branched compound containing two α2 , 3-linked Neu5Ac residues , one branching and one linear ( Fig . 1D ) . It is a carbohydrate sequence uniquely found on gangliosides . Tight crystal packing prevented us from obtaining a complex with the larger GT1b oligosaccharide , which had been shown to interact with MCPyV VP1 in vitro [19] . For simplicity , the ganglioside nomenclature will be used with respect to the GD1a and GT1b oligosaccharides from here on . In each complex , electron density was only observed for the Neu5Ac-α2 , 3-Gal motif that is common to all three investigated oligosaccharides and is also present in GT1b ( Fig . S1A–C ) . This disaccharide motif binds to a shallow binding site on the outer surface of VP1 , which is formed entirely by residues of the BC- , DE- and HI-loops of one VP1 monomer ( Fig . 1B , C ) . In all complexes , some binding sites are blocked by crystal contacts and thus not occupied , while others have bound the carbohydrate ligand , all in an identical manner ( Fig . 1B ) . There were some binding sites in each complex that were only weakly occupied , and into which the ligands were not modeled . In all instances where ligand was bound , the α2 , 3-glycosidic linkage between Neu5Ac and Gal adopts the same conformation ( torsion angles of −54° and −6° ) that is preferred in solution and that is also found in complexes of mPyV VP1 , hemagglutinins of influenza A viruses and wheat-germ agglutinin with linear α2 , 3-sialylated oligosaccharides [23] , [25] , [26] . The structure of MCPyV VP1 bound to the oligosaccharides is virtually identical to the unbound state , indicating that the carbohydrates dock into a preformed binding pocket . Neu5Ac forms the major contact point with MCPyV as it contributes most interactions and is best defined by electron density ( Fig . S1A–C ) . Most of its protruding functional groups are engaged by the protein . Its carboxyl group forms a salt bridge and hydrogen bonds with K299 and S297 in the HI-loop and water-mediated hydrogen bonds with D145 and S297 ( Fig . 1C ) . The Neu5Ac N-acetyl group faces away from the fivefold axis , interacting with residues in the BC-loop . It makes hydrophobic interactions with the side chains of W76 and Y81 and hydrogen bonds with D82 , which forms a salt bridge with K295 . Furthermore , the O4 hydroxyl group of Neu5Ac interacts with Y143 and D145 in the DE-loop via a water molecule . The glycerol chain of Neu5Ac faces away from the VP1 surface , and more than one conformation was observed for this chain in our complexes . Contacts with Neu5Ac are mostly mediated by side chains and all residues directly interacting with Neu5Ac are strictly conserved among MCPyV isolates and among newly identified MCPyV-like viruses of great apes [27] ( Fig . S2 ) . The Gal residue does not contact the protein directly and exhibits elevated temperature factors , but can be docked unambiguously into the electron density . It is likely stabilized by the conformational preferences of the glycosidic bond . Only the Neu5Ac-α2 , 3-Gal motif is clearly defined in our electron density maps , suggesting that this disaccharide unit serves as the main MCPyV binding target . In the DSL complex , only the internal Neu5Ac-α2 , 3-Gal sequence is contacted by the protein , while the terminal α2 , 8-linked Neu5Ac does not have clear electron density and is therefore flexible ( Fig . 1D , Fig . S1A ) . To probe the interaction of MCPyV VP1 with DSL in solution , we analyzed the complex using saturation transfer difference ( STD ) NMR spectroscopy . In this technique , saturation from a macromolecule is transferred to a small-molecule ligand , reaching only those parts of the ligand within roughly 5 Å from the protein . By inspecting the resulting STD spectrum , the macromolecule-bound parts of the ligand can be mapped . With the exception of the axial Neu5Ac proton H3 ( H3ax ) , all signals from the two Neu5Ac rings resonate at different frequencies and can therefore be clearly distinguished ( Fig . S3 ) . Saturation transfer from the protein was observed only to hydrogen atoms of the internal Neu5Ac and the connecting Gal , confirming the interactions seen in all crystal structures ( Fig . 2 ) . For the terminal Neu5Ac , the equatorial H3 ( H3eq ) proton as well as H4 , H5 and H6 protons receive no saturation , while all the equivalent protons for the internal Neu5Ac are observed in the STD spectrum ( Fig . S3 ) . The Neu5Ac H3ax signal in the STD spectrum most likely arises from contacts with the internal Neu5Ac only . Likewise , for the Neu5Ac methyl groups , significantly more transfer is observed for the internal Neu5Ac . The relatively small saturation of the terminal Neu5Ac methyl signal is also observed in the absence of protein and due to relaxation artifacts ( Fig . S3 ) . Some saturation transfer was also observed to the glycerol chain of the internal Neu5Ac ring as well as to some Gal protons ( Fig . 2 ) . These portions of DSL are not tightly tethered to the protein in the crystal structure , but they are within the 5 Å saturation transfer limit from the VP1 surface . We next asked whether MCPyV had a preference for one of the two Neu5Ac-α2 , 3-Gal motifs present in GD1a , the linear group on the “left arm” of GD1a or the branching group forming the “right arm” of the oligosaccharide ( Fig . 1D ) . Indeed , the electron density map of the GD1a complex contains weak electron density features close to the Gal residue , which are compatible with the linear Neu5Ac-α2 , 3-Gal epitope , but not with the branching one ( Fig . S1D ) . Moreover , addition of a branch to the MCPyV-bound Gal residue would result in steric clashes with the protein . Thus , we can unambiguously assign the MCPyV-binding epitope to the linear Neu5Ac-α2 , 3-Gal moiety on the “left arm” of GD1a . We then analyzed the MCPyV VP1 interaction with GT1b by STD NMR , and could confirm the interaction in solution ( Fig . S4 ) . Saturation transfer was observed for the methyl group of α2 , 3-linked Neu5Ac as well as for at least one Gal H1 proton , indicating that the Neu5Ac-α2 , 3-Gal epitope is likely also recognized on GT1b . However , heavy signal overlap in the heptasaccharide rendered further assignment difficult . Like GD1a , the GT1b oligosaccharide contains two Neu5Ac-α2 , 3-Gal motifs , one linear and one branching . Given the specificity of MCPyV for the linear epitope on GD1a , it is highly likely that MCPyV also binds the linear Neu5Ac-α2 , 3-Gal motif on GT1b . In summary , our epitope mapping establishes the linear Neu5Ac-α2 , 3-Gal disaccharide as the motif recognized by MCPyV VP1 , which is preferred over both a terminal α2 , 8-linked Neu5Ac residue and a branching Neu5Ac-α2 , 3-Gal motif . To probe the importance of sialic acid for MCPyV infection , we introduced mutations in the sialic acid binding site of MCPyV VP1 that either remove important interactions ( W76A , Y81V , K295A ) or create steric hindrance ( S297N ) . Western blots of mutant VP1 proteins expressed in mammalian cells revealed VP1 laddering , indicating the presence of disulfide crosslinks characteristic of assembled capsids ( Fig . 3D ) . Nuclease-digested purified stocks of wild-type and mutant capsids contained comparable amounts of encapsidated DNA , ranging from 0 . 12–0 . 14 ng of DNA per ng of VP1 . Thus , the mutations are unlikely to have caused major structural changes . Recombinant MCPyV VP1 has previously been shown to bind and hemagglutinate sheep RBCs by interacting with sialylated glycans [19] . The mutant capsids showed impaired hemagglutination ability , indicating that each mutated residue is functionally involved in forming the sialic acid binding site ( Fig . 3A ) . We then asked whether the mutant capsids are infectious . Pseudovirions generated using each of the mutant VP1s were deficient in infectious delivery of an encapsidated Gaussia luciferase reporter plasmid to cultured human A549 cells ( Fig . 3B ) , demonstrating a requirement for direct interactions between the MCPyV virion and sialylated glycans during the infectious entry process . A further set of experiments examining the binding of capsids to cells revealed that each of the VP1 mutant capsids bound to A549 cells at least as efficiently as wild type VP1 ( Fig . 3C ) . This observation is consistent with a prior report indicating that initial MCPyV-cell interactions are mediated primarily by non-sialylated GAGs [18] . The binding of each mutant was antagonized by pre-treatment of the cells with GAG-degrading enzymes ( Fig . 3D ) , confirming that attachment is mediated by GAGs , even for the mutant capsids . Pre-treatment of cells with GAG-degrading enzymes has previously been shown to decrease wild-type MCPyV infection due to a failure of the virus to stably attach to cells [18] . Interestingly , wild-type MCPyV showed weak residual binding to cells treated with GAG-degrading enzymes , while the mutants did not ( Fig . 3D ) . Thus , this low level of residual binding might arise from binding to sialylated oligosaccharides on host cells . As the sialic acid binding site mutants were capable of attaching to GAGs , our data demonstrate that the MCPyV GAG-binding motif is distinct from the sialic acid binding site . Sialic acid-containing oligosaccharides serve as receptors for related polyomaviruses such as mPyV , SV40 and JCPyV . Although all of these proteins recognize different sialylated ligands , sialic acid is the main point of attachment in each case , with auxiliary interactions determining the individual binding specificities [17] , [22] , [23] . Interestingly , there are three entirely different Neu5Ac binding modes among the four viruses , exemplified by MCPyV , mPyV and SV40 ( Fig . 4 ) . Such a high degree of variability is unusual . Related viruses usually possess virtually identical binding sites for sialic acid , as demonstrated by comparisons of JCPyV and SV40 [17] , [22] , [23] , or of different Influenza A viruses [28] . Specificity is in these cases achieved by augmenting contacts that surround the central , conserved sialic acid binding site and that modulate binding to different sialylated carbohydrates [17] , [22] , [23] . Interestingly , all but one of the MCPyV Neu5Ac binding residues are identical in similar viruses of great apes ( Fig . S2 ) , indicating that these viruses share a common sialic acid binding site . The Neu5Ac binding sites of MCPyV , SV40 and mPyV do not lie in different regions of VP1 , but rather use equivalent positions in sequence and in structure ( Fig . 4B–E ) . However , each binding site employs a unique set of residues in these positions that in each case engage Neu5Ac in a different orientation ( Table 2 ) . In the MCPyV complex , for instance , the N-acetyl group of Neu5Ac points away from the fivefold axis ( Fig . 4B ) , while it is oriented towards the clockwise neighboring monomer and inserts into a protein cavity in the SV40 complex ( Fig . 4C ) , and faces directly towards the clockwise neighboring monomer in the mPyV complex ( Fig . 4D ) . The N-acetyl group is contacted by a hydrogen bond to D82 and hydrophobic interactions with W76 and Y81 in the MCPyV complex , by a hydrogen bond to N272 and hydrophobic interactions with F270 and Q62 as well as F75* from the neighboring monomer in the SV40 complex , and by a hydrogen bond to Y72 in the mPyV complex ( Table 2 ) . Likewise , the carboxylate group of Neu5Ac is recognized by a hydrogen bond to S297 and a salt bridge to K299 in the MCPyV complex , by two hydrogen bonds to S274 and T276 in the SV40 complex and by a salt bridge to R78 in the mPyV complex ( Table 2 ) . Nevertheless , the Cα positions of the residues involved in Neu5Ac recognition are very similar in the three viruses , implying that the backbone structure in this region of VP1 is particularly suitable for evolving carbohydrate binding sites . Thus , while conserved amino acids imply that two proteins share a similar binding site , non-conserved amino acids do not exclude the possibility that two proteins engage similar carbohydrates in equivalent locations . The level of plasticity observed in the polyomavirus family could well be present in other families of carbohydrate-binding proteins where structural information is still lacking and does not yet allow for comparisons of binding modes .
We demonstrate here that the MCPyV major capsid protein VP1 directly interacts with carbohydrates bearing a linear Neu5Ac-α2 , 3-Gal motif . Our high-resolution structures of complexes reveal the molecular interactions governing recognition of this motif . The observed interaction has functional relevance as VP1 point mutants that lack sialic acid binding capability are unable to mediate infectious delivery of an encapsidated reporter plasmid to host cells . As these mutations did not affect GAG-dependent attachment , the sialic acid binding site described here must be functionally distinct from the as yet unidentified GAG-binding site on the virion surface . The idea that MCPyV infectious entry requires a direct interaction with a sialylated cellular glycan for an entry step that takes place after stable GAG-dependent attachment to the cell helps reconcile a prior report by Erickson and colleagues , who postulated that MCPyV infectious entry requires a direct binding interaction with sialylated glycans [19] , with a subsequent report by Schowalter and colleagues indicating that sialylated glycans are not required for MCPyV attachment to cultured cell lines [18] . A requirement for direct interactions between MCPyV VP1 and a sialylated glycan for a post-attachment infectious entry step also explains the past observation that Lec2 cells , which are deficient in biosynthesis of sialylated glycans , readily bind MCPyV but nevertheless do not support MCPyV infectious entry unless sialylated glycan synthesis is restored [18] . The treatment of cells with neuraminidases allows conclusions about viral entry in most cases . For example , neuraminidase treatment of cells dramatically reduces the infectivity of BKPyV , which uses gangliosides such as GT1b as sole receptors for both attachment and entry [15] . The infectivity of JCPyV , which uses the sialylated glycan LSTc as a primary receptor , is likewise sensitive to neuraminidase [17] , [29] . However , it can be difficult to determine , by neuraminidase treatment alone , whether viruses depend on sialylated glycans for entry . In fact , MCPyV infectivity does not appear to be affected by transient neuraminidase treatment [18] . An explanation for this finding is that MCPyV can use GAG-dependent binding to stably persist on the cell surface until neuraminidase activity wanes and newly synthesized sialylated glycans begin to reappear on the cell surface . In light of this explanation , it seems possible that sialic acid is used as a secondary receptor by other viral species , although such usage of sialylated glycans as post-attachment co-receptors is essentially unheard of among viruses investigated to date [30] . For some of these viruses , the need for a sialylated glycan co-receptor may have not yet been uncovered because engagement of the primary receptor influences the outcome of transient neuraminidase experiments , thus masking the involvement of sialic acids . Our data establish a linear Neu5Ac-α2 , 3-Gal disaccharide as a specific MCPyV binding motif . We have confirmed the interaction for the 3SLN , DSL , GD1a and GT1b oligosaccharides , which all contain this motif . Our data agree well with the observation that the ganglioside GT1b interacts with MCPyV VP1 pentamers in sucrose flotation assays [19] . However , it differs from the earlier observation that GD1a did not bind MCPyV VP1 [19] . We can exclude that GD1a binding in our experiments was mediated by crystal contacts as no symmetry-related VP1 molecules were bridged by GD1a . It is possible that the use of GST-tagged VP1 pentamers , which can form large aggregates by dimerization of GST-tags , interfered with GD1a binding in the earlier study . Linear Neu5Ac-α2 , 3-Gal disaccharides are present on many different classes of carbohydrates , such as gangliosides and other glycolipids , N- and O-linked glycoproteins [20] . As we do not see any contacts of MCPyV with carbohydrate residues outside the binding epitope in our structures , it is likely that MCPyV can bind to most oligosaccharides bearing a linear Neu5Ac-α2 , 3-Gal disaccharide . The MCPyV binding epitope is thus smaller and present on more oligosaccharides than other sialylated polyomavirus ligands , such as the SV40 receptor GM1 and the JCPyV receptor LSTc , which are both recognized with higher specificity [17] , [22] . In contrast , MCPyV likely binds sialylated oligosaccharides in a more promiscuous manner , similar to the well-characterized mPyV , whose receptor interactions have interesting parallels to MCPyV . First , the mPyV binding epitope is also the linear Neu5Ac-α2 , 3-Gal disaccharide [23] , [31] , and second , mPyV can bind to several different oligosaccharides bearing that motif . Notably , only few of them , the gangliosides GD1a and GT1b , are known to mediate mPyV infection [16] , [32] , [33] , while other carbohydrates present on glycoproteins are hypothesized to be ‘pseudoreceptors’ for mPyV that bind the virus but do not mediate infection [33] , [34] . Gangliosides were therefore early MCPyV receptor candidates , especially because the ganglioside GT1b was known to interact with MCPyV in vitro [19] . However , GT1b supplementation of ganglioside- or sialyl glycan-deficient cells did not rescue MCPyV infection , while it did rescue the infectivity of BKPyV [18] ( RMS and CBB , unpublished data ) . Thus , GT1b or other gangliosides are unlikely to serve as functional secondary receptors for MCPyV . We think it likely that MCPyV is able to bind many sialylated oligosaccharides on host cells and that , similar to mPyV , both functional and pseudoreceptors regulate entry processes . However , further studies will be needed to elucidate the roles of the various glycans bearing linear Neu5Ac-α2 , 3-Gal disaccharides in MCPyV infection , and to define which are the functional secondary receptors and which act as pseudoreceptors . This model does not exclude the possibility that MCPyV might engage an as yet unidentified longer oligosaccharide bearing a binding epitope with additional contacts and therefore serving as a higher affinity secondary receptor . However , the examples of mPyV and of human Influenza A viruses show that despite promiscuous binding properties , a higher affinity receptor may not be required [28] , [33] . Taken together with previous studies , our findings strongly support a novel uptake pathway in which infectious entry of MCPyV requires both initial attachment to GAGs and subsequent interaction with a sialylated oligosaccharide . Our structural and functional data allow conclusions about several of the steps involved . Entry is initiated by attachment of MCPyV to GAGs on the cell surface , which is likely mediated by the major capsid protein VP1 as there is no indication in polyomavirus structures of exposed minor capsid proteins [31] , [35] . Since our mutant pseudovirions deficient in sialic acid binding still were capable of GAG-mediated attachment , the two binding sites are separated entities on the VP1 surface , and sialic acids and GAGs do not use a dual function binding site . However , our structural data do not allow prediction of the GAG binding site as there is no conservation of GAG binding among polyomaviruses . Attachment to GAGs does not seem to be a prerequisite for sialic acid binding , as the protein clearly bound sialylated ligands in the absence of GAGs ( Figs . 1 , 2 ) . Instead , we think it likely that the initial attachment to GAGs helps to concentrate viral particles at or close to the cell surface , perhaps compensating for the relatively low affinity of the MCPyV VP1-sialic acid interaction . The KD value of this interaction must be in the mM range because the STD NMR experiment only covers µM-mM interactions [36] , and oligosaccharide concentrations in the mM range were necessary to obtain complex crystals . While the MCPyV-GAG interaction is well characterized on the functional level , but not structurally , it is the other way round for the interaction with sialic acid . Two key questions remain . First , which of the many oligosaccharide moieties bearing the epitope described here is the functional secondary receptor for MCPyV ? And second , what is its role during MCPyV entry ? One enticing possibility is that the sialic acid-dependent step in MCPyV entry mediates intracellular trafficking . Different polyomaviruses use differing machineries for initial uptake , such as cholesterol-mediated endocytosis for SV40 , BKPyV and mPyV , or clathrin-dependent endocytosis in the case of JCPyV [37] , [38] . However , trafficking after uptake appears to converge en route to the ER , and to depend on sialylated glycans at least in some cases [16] , [33] , [39] , [40] . MCPyV might use similar trafficking routes after GAG-dependent attachment . However , more functional studies will be necessary to define its role , and the point mutants in the sialic acid binding site we describe here might be interesting tools in this investigation . Importantly , the MCPyV uptake pathway differs from those of other polyomaviruses investigated to date , none of which require GAGs for infection [15] , [16] , [29] , and from papillomavirus entry pathways , which depend on GAGs , but do not need sialic acid [14] . It also differs from the receptor requirements of some adeno-associated viruses that are able to bind both GAGs and sialic acids , but can use either one as receptors [41] , and do not require ordered , direct interactions with both . The present findings may also inform the study of other viruses because it is quite likely that MCPyV is not the only virus to rely on both GAGs and sialic acid to infect cells . In conclusion , we have established a specific interaction of MCPyV VP1 with a linear sialylated disaccharide , and demonstrate the functional relevance of this interaction for MCPyV infection . The observed interactions provide a useful platform for the development of MCPyV-specific vaccines and antivirals . The novel uptake mechanism of MCPyV , requiring GAGs and sialic acid sequentially , furthermore enhances understanding of cell entry by carbohydrate-engaging viruses . Comparison with other polyomavirus-receptor structures demonstrates the high level of adaptation that sialic acid binding sites can undergo , informing both viral and non-viral protein-carbohydrate recognition processes .
Two different GT1b oligosaccharide compounds with different linkers on the terminal glucose were used in this study for crystallization and NMR experiments . The first compound ( 1 ) was synthesized as described [42] and carries a CH2-CH2-CH2-N3 linker at the anomeric carbon of the Glc ring . The second compound ( 2 ) was synthesized from the lactose derivative 2- ( trimethylsilyl ) ethyl 2 , 6-di-O-benzyl-β-D-galactopyranosyl- ( 1→4 ) -2 , 3 , 6-tri-O-benzyl-β-D-glucopyranoside using the reaction scheme described by Ishida et al . [43] , [44] . It carries a CH2-CH2-Si ( CH3 ) 3 linker on the anomeric carbon of the Glc ring . In both compounds , the linkers inhibit mutarotation between α- and β-Glc . GD1a oligosaccharide was produced by ozonolysis of GD1a ganglioside as described [45] , [46] . DNA coding for amino acids 38–320 of w162 MCPyV VP1 ( GenBank # FJ392560 ) was amplified by PCR and cloned into the pET15b expression vector ( Novagen ) in frame with an N-terminal hexahistidine tag ( His-tag ) and a thrombin cleavage site . The protein was overexpressed in E . coli BL21 ( DE3 ) and purified by nickel affinity chromatography and gel filtration on Superdex-200 ( GE Healthcare ) . The tag was cleaved with thrombin prior to gel filtration , leaving the non-native amino acids GSHMLE at the N-terminus . MCPyV VP1 was in a buffer comprised of 20 mM HEPES pH 7 . 5 , 150 mM NaCl and 20 mM DTT after gel filtration . The protein was concentrated to 3 . 5 mg/mL and crystallized at 20°C by hanging drop vapor diffusion against a reservoir solution containing 100 mM sodium cacodylate pH 6 . 5 , 6% ( w/v ) PEG 3 , 350 and 300 mM magnesium chloride . A seeding stock was included in the crystallization drops . Crystals were harvested into reservoir solution , cryoprotected by soaking them in reservoir solution supplemented with 25% ( v/v ) glycerol for 10 s , and flash-frozen in liquid nitrogen . For oligosaccharide complex formation , crystals were soaked in reservoir solution supplemented with 20 mM DSL ( Sigma ) , 25 mM GD1a oligosaccharide or 20 mM 3SLN ( Sigma ) for 10–40 min . The same concentration of oligosaccharide was also included in the cryoprotection solution . Similar approaches to cocrystallization were used for unsuccessful attempts to obtain complexes with GT1b oligosaccharides . All diffraction data were collected at −180°C and at a wavelength of 1 Å . Datasets for native and complexed MCPyV VP1 were recorded at beamlines 14 . 1 at BESSY ( Berlin , D ) and X06DA at SLS ( Villigen , CH ) , respectively ( Table 1 ) . X-ray data were processed with xds [47] , and the structure was solved by molecular replacement with Molrep [48] using a search model generated from the structure of mPyV VP1 ( pdb 1VPN ) [23] . Each structure contained four VP1 pentamers in the asymmetric unit . The structures were completed by iterating rounds of model building in Coot , and with simulated annealing , restrained coordinate and B-factor as well as TLS refinement in Refmac5 and Phenix [49] , [50] , [51] . Initial TLS parameters were generated using the TLSMD server [52] . The 20-fold non-crystallographic symmetry linking the VP1 monomers in the asymmetric unit was used as a restraint throughout refinement for protein regions outside crystal contacts . Oligosaccharide ligands were located in weighted mFo-DFc difference electron density maps , and refined using restraints from the Refmac library and user-defined ones for the glycosidic linkages of sialic acid . The final models agree well with the experimental data and have good geometry ( Table 1 ) . In all models , more than 95 . 5% of amino acids are in the favoured region of the Ramachandran plot , with <0 . 1% in the disallowed regions . Coordinates and structure factor amplitudes for the native , DSL complex , 3SLN complex and GD1a complex MCPyV VP1 structures were deposited in the PDB under accession codes 4FMG , 4FMH , 4FMI , and 4FMJ , respectively . NMR spectra were recorded at 283 K using 3 mm tubes on a Bruker DRX 500 MHz spectrometer fitted with a 5 mm cryogenic probe ( DSL interaction ) or on a Bruker AVIII-600 spectrometer equipped with a room temperature probehead ( GT1b interaction ) , and processed with TOPSPIN 2 . 0 ( Bruker ) . A sample containing 42 µM MCPyV VP1 , 1 mM DSL ( Sigma ) , 20 mM deutero-Tris pH 7 . 5 , 20 mM deutero-β-mercaptoethanol , 150 mM NaCl was used for the STD NMR analysis of the DSL-VP1 interaction . For the GT1b-VP1 interaction , a sample containing 26 µM MCPyV VP1 and 2 mM GT1b was prepared using the same buffer as for the DSL-containing sample . Samples containing 1 mM DSL or 2 mM GT1b but no protein were prepared and were used for spectral assignment and to confirm that the chosen on-resonance frequencies did not directly excite the ligands . Samples were prepared in D2O and no additional water suppression was used in order not to affect the anomeric proton signals . The off- and on-resonance frequencies were set to 80 ppm and 7 ppm , respectively . The total relaxation delay was 4 s . A cascade of 40 Gaussian-shaped pulses with 50 ms duration each , corresponding to a strength of 65 Hz , and a saturation time of 2 s was used for selective excitation . A 10 ms continuous-wave spin lock filter with a strength of 3 . 7 kHz was employed in order to suppress residual protein signals . 32 k points were collected and zero filling to 64 k data points was employed . Spectra were multiplied with an exponential line broadening factor of 2 Hz prior to Fourier transformation . Spectra were referenced using HDO as an internal standard [53] . Pure oligosaccharide samples containing 1 mM DSL or 2 mM GT1b served as samples for spectral assignment . Series of 1D 1H-TOCSY and COSY spectra as well as 1H , 13C-HSQC spectra were acquired for assignment of the oligosaccharide proton resonances . Literature values on related oligosaccharides served as assignment controls [54] , [55] , [56] . A previously-reported system for production of MCPyV-based reporter vectors ( pseudoviruses ) [57] was used for functional analysis of mutant VP1 proteins . Briefly , expression constructs carrying codon-modified ORFs encoding MCPyV VP1 and VP2 proteins were co-transfected into 293TT cells along with a reporter plasmid encoding Gaussia luciferase . The resulting MCPyV pseudovirions were released from the transfected cells by detergent lysis in the presence of a DNase/RNase cocktail . The pseudovirions were allowed to mature overnight , then purified by ultracentrifugation through Optiprep gradients . Gradient fractions were screened for the presence of encapsidated DNA using Picogreen reagent ( Invitrogen ) . VP1 content of the purified reporter vector stocks was standardized based on SYPRO Ruby ( Invitrogen ) stained SDS-PAGE gel analysis . HA assays , infectivity assays and cell binding assays as well as heparinase/chondroitinase treatment were carried out as previously reported [18] .
|
Viruses must interact with specific receptor molecules on their host cells in order to first attach to the cell and second gain entry into it . Therefore , a viral entry pathway is a sequence of precisely regulated binding events between viral proteins and their cellular receptors , which can be proteins or other biomolecules . In the present study , we investigated the human Merkel cell polyomavirus ( MCPyV or MCV ) and show that it uses complex carbohydrates containing sialic acid as receptors for entry . MCPyV was discovered in 2008 , is widespread in humans and can cause aggressive skin tumors termed Merkel cell carcinomas in immunosuppressed individuals . We determined the crystal structures of the MCPyV capsid protein bound to sialylated carbohydrates , describing the contacts needed for receptor recognition in molecular detail . When we introduced targeted mutations that abolished sialic acid binding into the virus , it was unable to infect cells although it could still attach to them . Earlier studies showed that the virus uses a different group of carbohydrates called glycosaminoglycans for initial attachment to the cell surface . Thus , its entry pathway involves sequential binding to two distinct classes of carbohydrates . Our structures can be used as a starting point to develop antivirals against MCPyV .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biomacromolecule-ligand",
"interactions",
"medicine",
"viruses",
"and",
"cancer",
"emerging",
"viral",
"diseases",
"microbiology",
"viral",
"structure",
"oncology",
"oncology",
"agents",
"proteins",
"biology",
"biochemistry",
"virology",
"antivirals",
"glycobiology"
] |
2012
|
Structures of Merkel Cell Polyomavirus VP1 Complexes Define a Sialic Acid Binding Site Required for Infection
|
Our 13C- and 1H-chemical exchange saturation transfer ( CEST ) experiments previously revealed a dynamic exchange between partially closed and open conformations of the SAM-II riboswitch in the absence of ligand . Here , all-atom structure-based molecular simulations , with the electrostatic effects of Manning counter-ion condensation and explicit magnesium ions are employed to calculate the folding free energy landscape of the SAM-II riboswitch . We use this analysis to predict that magnesium ions remodel the landscape , shifting the equilibrium away from the extended , partially unfolded state towards a compact , pre-organized conformation that resembles the ligand-bound state . Our CEST and SAXS experiments , at different magnesium ion concentrations , quantitatively confirm our simulation results , demonstrating that magnesium ions induce collapse and pre-organization . Agreement between theory and experiment bolsters microscopic interpretation of our simulations , which shows that triplex formation between helix P2b and loop L1 is highly sensitive to magnesium and plays a key role in pre-organization . Pre-organization of the SAM-II riboswitch allows rapid detection of ligand with high selectivity , which is important for biological function .
Non-coding RNAs are currently thought to account for over 75% of the human genome [1] . In bacteria , non-coding RNAs play important roles in gene regulation . One such class of RNAs , riboswitches , regulates metabolite production . Here , a single RNA sequence folds into one of two or more mutually exclusive folds depending on the metabolite concentration [2 , 3] . In some cases , such as the S-adenosylmethionine-I ( SAM-I ) riboswitch , the RNA contains a transcriptional terminator that forms when ligand is present , in effect silencing genes important for ligand production [4–6] . When the ligand is not present , the terminator does not form , allowing gene expression , and therefore ligand production , to continue efficiently . In other cases , such as the SAM-II riboswitch , ligand binding may lead to sequestration of the Shine/Dalgarno sequence , likely blocking ribosome binding and , as a consequence , protein synthesis [7] . While these examples of ligand-dependent secondary structure switches have been known for some time , a detailed thermodynamic understanding at the atomistic level , including the indispensable effect of the RNA’s ion-atmosphere , has not been achieved . In recent years , riboswitches have become canonical systems for studies of diverse RNA behaviors , as they possess quintessential characteristics of many RNA systems: ligand binding , Magnesium ion ( Mg2+ ) sensitivity , conformational changes , secondary structure remodeling , and regulatory functions . Chemical footprinting , NMR , small-angle X-ray scattering ( SAXS ) and single molecule FRET techniques are being exploited to elucidate the folding kinetics , thermodynamics and the magnesium ion sensitivity in RNA systems such as the TPP riboswitch [8] , glycine-dependent riboswitches [9] , different variants of P4-P6 RNA [10–12] and P5abc subdomain of the Tetrahymena group I intron ribozyme [13] . Other work focuses on more complex functions , such as splicing and ligand recognition and their associations with proteins or different metabolites [14 , 15] . Success in understanding the structural , dynamical and functional aspects of riboswitch systems requires an integrated experimental and theoretical approach . Traditional crystallographic techniques produce static snapshots of the riboswitch . SmFRET , NMR , and SAXS methods obtain kinetic information and overall distributions of conformations . Molecular simulation allows one to integrate disparate experimental data into a single coherent picture , characterizing transitions in atomistic detail and the free energy landscape with fine resolution . A large number of riboswitches have been crystallized and have also been investigated via fluorescence and single molecule techniques [16–28] . Molecular simulations have also been used to study a number of riboswitches , including but not limited to the SAM-I , SAM-II , pre-Q , and adenine riboswitches [29–32] . Some of these studies replaced the essential atmosphere of divalent Mg2+ ions with other , monovalent ions . Others used a repulsive Debye-Hückel interaction between the phosphate groups of the RNA . Such implicit treatments of the ions , however , neglect important near-field effects that occur inside the core of the riboswitch , where the ions are strongly coupled to the RNA . Although the role of Mg2+ in stabilizing the RNA tertiary structure has long been realized [33 , 34] , the molecular basis of ion-RNA interactions , in terms of structure and function , is not well understood . In a pioneering study , Draper and co-workers distinguished three classes of ion environments: ( i ) the diffuse ions , which are not restrained to any particular region , ( ii ) water-surrounded ions separated from the RNA by a single hydration layer , which we call outer-sphere ions , and ( iii ) chelated ions in the inner sphere , which form direct contacts with at least two different phosphate groups of the RNA [34 , 35] . While the potential role of chelated ions has been emphasized in many studies , recent work using explicit solvent molecular dynamics instead highlights a dense layer of outer-sphere Mg2+ ions , which are primarily responsible for anchoring the RNA structure [36] . These outer-sphere Mg2+ ions are only transiently bound but nonetheless strongly coupled to the RNA dynamics . Such highly correlated Mg2+ ions may even reside in the core region of riboswitch RNAs [36 , 37] . This dynamic cloud of Mg2+ has also been simulated in all-atom reduced models that combine Manning theory with a background of monovalent ions , represented by Debye-Hückel interactions [37 , 38] . In addition to these native basin simulations , studies of metabolite recognition and specificity have also been initiated using conformational ensemble sampling , again in the absence of Mg2+ ions [39] . While most riboswitch studies have focused on riboswitches in the 5’-UTR of mRNA that control transcription , less attention has been paid to translational control by riboswitches through ligand-dependent sequestration of the ribosome binding site ( i . e . , the Shine/Dalgarno sequence ) . The SAM-II riboswitch is one such relatively small RNA element which regulates methionine and SAM biosynthesis . A single hairpin , classic H-type pseudoknot and triplex interaction near the ligand binding site make this RNA an interesting system to study RNA control over the translation initiation process [21 , 40–44] . A previous single molecule fluorescence resonance energy transfer ( smFRET ) [21] imaging study sheds light on the dynamic nature of ligand-free SAM-II riboswitch , which becomes conformationally restrained upon ligand binding . The flexibility of such highly transient conformations is tuned to ensure a viable time scale for conformational transitions in the absence of ligand . More rapid sensing could , however , be achieved if the riboswitch adopted a binding competent conformation in the ligand-free state . Mg2+ ions can act as effective anchors , aiding in the preservation of the structural integrity of the RNA . The emergence of two distinct FRET configurations in the presence of 2 mM Mg2+ in the ligand free system suggests that Mg2+ has the ability to compress the RNA structure , in such a way that it might pre-organize the RNA to form a binding competent conformation . In a series of small angle X-ray scattering ( SAXS ) experiments , we observed an analogous signature of Mg2+ induced structural collapse that can facilitate subsequent ligand binding [40] . Our studies provided direct insight into the global rearrangement induced by both Mg2+ and ligand . The compaction of RNA by Mg2+ was also studied by size-exclusion chromatography ( SEC ) : changes in the measured elution volume suggested a decrease in the particles’ hydrodynamic radius [40] . Pre-organization by Mg2+ has also been observed in other riboswitches [21 , 28 , 45] . In the SAM-I system , our biochemical studies have shown that addition of Mg2+ yields the pre-organized partially folded state . In addition , we have shown that , in the absence of Mg2+ , the fully folded state cannot be achieved , even at high ligand concentrations [23 , 28 , 46] . The presence of both , Mg2+ and ligand are required for the stabilization of the fully folded ligand-bound configuration . While previous smFRET and SAXS data revealed that ligand free RNA undergoes substantial structural changes upon variation of Mg2+ concentration , these structural changes often remained undetected by traditional NMR and X-ray crystallography techniques because of the transient nature and low population levels of such intermediates [47] . The newly developed Chemical Exchange Saturation Transfer ( CEST ) measurements are now capable of probing these sparsely and transiently populated RNA conformations . Earlier we studied NMR dynamics of the SAM-II system with this new method [47] . The data indeed confirmed that SAM-II riboswitch can access a sparsely populated but bound-like pre-organized state even in the absence of ligand [40 , 47] . In the present study , we performed molecular simulations to predict the effect of Mg2+ on the conformational landscape of the SAM-II riboswitch . We then tested these predictions with 13C-CEST data . Analysis of our simulations yields the free energy landscape of the SAM-II riboswitch , the effect of Mg2+ on this landscape and insight into the microscopic origins of these effects . More specifically , reappraisal of the 13C-CEST data for the ligand-free SAM-II riboswitch at different Mg2+ concentrations enabled us to probe the influence of Mg2+ on sparsely populated bound-like pre-organized states . We then revisited earlier smFRET , SAXS and SEC elution profiles and compared with our present equilibrium simulation results to integrate these data into a unified scenario of Mg2+-induced collapse . We calculated the free energy landscape of the SAM-II riboswitch using our recently developed all-atom structure-based model ( SBM ) that includes explicit Mg2+ ions and the effects of Manning condensation and Debye-Hückel Potassium and Chloride interactions . We specifically predict that , as Mg2+ concentration is increased from 0 . 25 mM to 2 mM , the SAM-II riboswitch collapses from an extended , partially unfolded state to a highly compact , pre-organized state , in agreement with the 13C-CEST studies , where we observe a shift in population towards a bound-like conformation . In addition , our simulations characterize this collapse transition in terms of the radius of gyration as a function of Mg2+ concentration , which is qualitatively similar to previous SAXS measurements . This agreement gives us confidence in the microscopic details of our simulations , showing that the triplex formation between helix P2b and loop L1 plays an important role in the collapse process .
As mentioned earlier , CEST experiments are able to capture transiently populated dynamic conformations [13] . This strategy was applied to the ligand-free SAM-II riboswitch in the presence of 0 . 25 mM and 2 mM Mg2+ . The 13C-CEST profiles of the labeled ribose C1’ and base C6 carbons of C43 were recorded at three different B1-fields ( 17 . 5 , 27 . 9 , 37 . 8 Hz ) with a mixing time of 0 . 3 s at 298 K [47] . We compare the data for 0 . 25 mM and 2 mM Mg2+ concentrations at B1-field of 17 . 5 Hz ( Fig 2a ) . The data were fit with a two-state model where the low population of the partially closed state ( peaks around 300 Hz spin-lock offset ) appears to increase with addition of 2 mM Mg2+ . Consistent results were obtained from CEST profiles for other B1-fields of 27 . 9 and 37 . 8 Hz ( Fig S1 in the S1 Text ) . Trajectory plots of the fraction of native contacts extracted from the generalized Manning equilibrium simulations of ligand-free SAM-II at these two concentrations clearly show the hopping between different conformations ( Fig 2b ) . Furthermore , the dynamic transitions between the two major states ( bound-like: Q≈0 . 9 and open: Q≈0 . 7 ) visit native-like conformations ( Fig 2c ) more frequently at 2 mM than at 0 . 25 mM Mg2+ , as summarized in the corresponding contact histograms , P ( Q ) ( Fig 2d ) . Both CEST experiments and simulation data indicate that the equilibrium shifts from the open conformations toward the native bound-like state as we increase Mg2+ concentration . The signature of the existence of such Mg2+ induced bound-like states has also been reported in previous smFRET experiments ( Fig 2e ) [22] . To support our observations we have revisited some of these smFRET efficiency assessments [22] and compared them with theoretical FRET predictions obtained from our generalized Manning model simulations under similar buffer conditions . The equation used for theoretical FRET prediction is described in section S1 in the S1 Text . We tracked the dynamics of positions 14 and 52 , where acceptor ( cy5 ) and donor ( cy3 ) fluorophore labels were placed in the smFRET experiments ( Fig 2f ) . Both experimental and simulation FRET confirm the coexistence of two states at 2 mM Mg2+ ( Fig 2g ) [22] . Previous SAXS data corroborates well the existing smFRET observations [22 , 40] . The SAXS data also indicated both ligand and Mg2+ ions are required to effectively fold this riboswitch . To microscopically understand their mutual and stand-alone effects from the present simulations , we studied the conformational differences of this riboswitch in four extreme buffer conditions and compared our computational results with experimental SAXS data . For this comparison , we extracted multiple snapshots from several long trajectories and computed ensemble averaged SAXS profiles using the Debye formula for spherical scatterers parameterized in the FoXS web server [48 , 49] as described in section S2 in the S1 Text . The predicted SAXS curves here show qualitative agreement with experiments ( Fig 2h ) [40] . The Kratky representation of SAXS data presented in Fig s2 in the S1 Text shows a pronounced peak , indicating the emergence of more extended conformations with decreasing Mg2+ concentrations . We note that capturing the entire conformational heterogeneity of an extended state is computationally challenging . This mostly applies for the extreme case where neither ligand nor Mg2+ is present . In this case , the correlation between theoretical and experimental SAXS profiles leaves room for improvement . Values for chi-square reflect that and are shown in Table S1 in the S1 Text . These analyses indeed suggest the potential impact of both , ligand and Mg2+ , stabilizing the closed conformations , which we characterize further below with contact data to describe the pre-organization and the ligand-organized closing . A significant Mg2+ induced collapse transition , as indicated by the SAXS data , has been followed over a wide concentration range ( up to 100 mM ) of Mg2+ in SEC elution volume profile ( Fig 2i ) . Here RNA elutes after longer retention times with increasing Mg2+ concentration ( [Mg2+] ) in the mobile phase [40] . Bigger elution volume signifies decreasing hydrodynamic radius of a monomeric RNA molecule . The folding transitions , both from experimental elution volume data and from average Rg measured from the equilibrium simulation analysis as functions of [Mg2+] , follow sigmoid curves with transition midpoint , Mg1/2 at 6 mM ( Fig 2i ) . At this point , a range of experimental techniques and simulation data support the existence of pre-organized states . Here we aim to obtain a thermodynamic description of how Mg2+ governs the energy landscape of RNA from our model simulation study . In Fig 3a , we show the free energy landscape for the folding transition of SAM-II riboswitch in SAM-bound ( in the presence of explicit ligand ) and SAM-free ( in the absence of ligand ) conditions near the physiological concentration of Mg2+ ( [Mg2+] = 2 . 0 mM ) . During this folding transition , each secondary structural segment folds sequentially illuminating the pathway of folding ( Fig 3b ) . The free energy profile , in the presence of explicit SAM has a distinct bound-state-well , reflecting the ligand-induced stabilization of the closed conformations ( designated as ( i ) in Fig 3c ) . In the apo-form of the riboswitch , the fully closed bound state does not correspond to a minimum in the landscape . At lower Q than this ligand-bound state , the free energy profile for apo-SAM-II riboswitch reveals three distinct minima . They involve: a ligand-free partially closed state ( state ( ii ) in Fig 3c ) , which has a substantial overlap with the ligand-bound closed conformation . In this state , the nonlocal contacts ( involving base-pairing contacts ) including base-stacking contacts in P1 , the P1-L3 pseudo-knot interaction , and major segments of P2b and the L1-P2b triplex interactions remain secured , while the contacts involved in Shine/Dalgarno sequence ( AAAG50G51A523´ ) , and in the part of L1-P2b are disrupted ( state ( ii ) in Fig 3c ) . Recent fluorescence and NMR spectroscopic data also indicated that C16 in P2a helix remains mostly unpaired in the absence of SAM [22] . The data also suggested that formation of the pseudoknot in the absence of SAM is highly transient in nature . Intermediate states , ( iii ) and ( iv ) in Fig 3c , although marginally separated by a small barrier , effectively belong to a broad , flat basin which involves an ensemble of partially folded open configurations . A representative unfolded structure ( ( v ) in Fig 3c ) is shown to describe the unfolded minimum . As we increase the concentration of Mg2+ we find enhanced stabilization of the pre-organized partially closed conformations ( state ( ii ) in Fig 3d ) relative to the open conformations . Our latest 13C-CEST chemical exchange data anticipated that the emergence of Mg-induced pre-organization can have immense consequences for rapid ligand recognition [47] . In the context of the simulation , Mg2+ induced thermodynamic stabilization is reflected by the difference in stability , ΔGPC-PO between the bound-like partially closed ( PC ) conformation and the partially open ( PO ) conformation and by ΔGU-O between the unfolded ( U ) and the open conformation ( O ) . These two stability differences , ΔGPC-PO and ΔGU-O vary upon increasing [Mg2+] until they reach their saturation limits . Both ΔGPC-PO and ΔGU-O plotted as functions of [Mg2+] , are fitted well to sigmoid curves with a Mg2+1/2 value around 6 mM ( inset of Fig 3d ) which again correlates well with the SEC elution volume data ( Fig 2i ) [40] . To address the open question of how Mg2+ ions regulate structural collapse , we have determined the Mg2+ distribution in the ion-solvation layer of SAM-II , which accommodates increasing numbers of Mg2+ up to 8 mM Mg2+ content ( Fig 4a ) . Subsequent additions of Mg2+ beyond 8 mM do not effectively add to the 1st layer of Mg2+ solvation . How we characterize the ion-solvation layer from our simulated trajectories is described in section S3 in the S1 Text ( Fig s3 in S1 Text ) . We have further classified the outer sphere Mg2+ present in the ion-solvation layer into two categories based on their number of associated phosphate groups: ( i ) Single phosphate coordinating Mg2+ ( Fig 4b ) , which efficiently neutralize the negative charge of the adjacent phosphate ( Fig 4d ) , and ( ii ) multiple phosphate coordinating Mg2+ ( Fig 4c ) . The key role in stabilizing the structure is played by such Mg2+ bridging multiple phosphates , which can act as glue in compact structures by holding a number of negatively charged phosphates together in close proximity . The population shift coincident with multiple coordinated Mg2+ ions with increasing Mg2+ concentration directly supports their role in stabilizing the structure ( Fig 4e ) . We have also investigated the thermodynamic impact of Mg2+-mediated phosphate contacts ( PHOSCont: total number of pair-wise phosphate-phosphate contacts ) on the energy landscape as a function of overall folding progress , expressed by the number of native contacts ( NCont ) , as shown in Fig 4f–4h . As we increase the Mg2+ concentration the broad minimum that appeared around NCont~800 , involving partially folded open conformations , gradually becomes more stabilized . Concurrent enrichment of phosphate-phosphate contacts extends the contour of the minimum asymmetrically toward higher PHOSCont . Additionally , by 8 mM [Mg2+] , the bound-like pre-organized state grows with substantial population , stabilized again by phosphate connections ( Fig 4h ) . We analyzed long equilibrium trajectories of the apo- and bound-forms of SAM-II slightly below the folding temperature in order to capture the essential characteristics of the pre-organized state , and also to compare this state with the fully folded ligand bound state . We have evaluated the distribution of native contact formation in each segment of secondary structure as a function of the total Q at different Mg2+ concentrations . Plots show contact formation in P2b ( Fig 5a–5d ) and the triplex interaction between helix P2b and loop L1 ( Fig 5e–5h ) , which are most affected by Mg2+ concentration . Data for the nonlocal contacts of P1 , L3-P1 , P2a , which appear only marginally affected by Mg2+ concentration , are shown in Fig s4 in S1 Text . The two distinct basins visible at low [Mg2+] , for the P2b helix and L1-P2b triplex contacts correspond to the pre-organized ( at higher Q ) and open states ( at lower Q ) . At increasing [Mg2+] , the populations gradually shift towards the pre-organized state . Around 8 mM Mg2+ , the dominant contribution arising from this pre-organized triplex to the conformational space is evident from Fig 5c and 5g . Ligand binding also strongly favors structure formation , even at moderate [Mg2+] , as the ligand bridges the gap between L1 strand and P2b helix , producing the fully formed triplex .
Motivated by our 13C-CEST profiles for SAM-II and their [Mg2+] dependence we have explored the free energy landscape of the SAM-II riboswitch using a recently developed all-atom SBM that includes explicit Mg2+ ions , Debye-Hückel treatment of implicit KCl interactions , and the effects of Manning condensation to accurately account for the ion atmosphere around the RNA . Our results support a mechanism involving Mg2+ induced pre-organization followed by conformational selection by the ligand , SAM , as we speculated in an early study [47] . The free energy analysis validates the observations of that pre-organization , providing an atomistic and thermodynamic basis for the enhanced population of a partially collapsed , pre-organized ensemble at sufficiently high Mg2+ concentration in the absence of ligand . We observe three distinct sets of conformations in the folding free energy landscape of ligand-free SAM-II riboswitch: ( i ) an ensemble of unfolded conformations , ( ii ) a broad ensemble of partially folded open conformations , and ( iii ) an ensemble of pre-organized bound-like conformations . As we increase magnesium concentration beyond 2–4 mM , the bound-like ensemble is further stabilized , shifting the equilibrium toward the pre-organized states . All the experimental results from our 13C-CEST profile , recent SAXS , single molecule FRET , and size-exclusion chromatographic studies are assembled and found to be in good agreement with the present simulation results ( Fig 2 ) . At higher concentrations , Mg2+ stabilizes compact structures by coordinating multiple charged phosphate groups of RNA in close proximity . The experimental results , together with free energy landscapes confirm that sufficient Mg2+ can indeed promote stable ligand binding in the SAM-II riboswitch , and is likely the structural basis for the switching control of protein translation . While this structural pre-organization of SAM-II can assist in rapid ligand recognition , our study suggests that a sufficiently high concentration of Mg2+ is necessary to capture those pre-organized states . Only when the system achieves a well-organized ion solvation layer at high [Mg2+] , the effect of additional Mg2+ seems limited . This layer involves a number of Mg2+ ions , each coordinating with multiple phosphate groups . Mg2+ ions thus serve as glue to the negatively charged phosphates and facilitate the structural compaction . We note that while chelated Mg2+ may play an important role in other riboswitch RNAs , no specific chelated ions have been reported so far in the SAM-II system . Our molecular simulation trajectories also allow us to pinpoint the structural basis of the effect , revealing that triplex interaction between the helix P2b and its association with the L1 strand dominate the process of pre-organization as summarized in Fig 5a–5c and 5e–5g , showing the gain of structure with increasing [Mg2+] . In the final step , ligand binding firmly bridges the extended gap between L1 and P2b , which seems otherwise not achievable through the addition of small , dynamic Mg2+ alone . But although P2b and its connection with L1 can be secured by the ligand , its presence again alone cannot fully stabilize the overall structure without addition of significant amount of Mg2+ ( Fig 2h ) . These findings suggest that a sufficiently high concentration of Mg2+ is necessary to stabilize the pre-organized triplex and then the presence ligand promotes the native triplex formation , as summarized in Fig 5d and 5h . We note that triplexes have recently emerged as important players in gene regulation by non-coding RNAs [50–53] . Base triples also play a role in RNase P and the Diels-Alder ribozyme [54] . Heroic calculations , as such recent microsecond explicit solvent simulations of riboswitches , will also shed light on these effects , especially regarding the role of solvation [55] . Nucleic acid-ion interactions make a substantial energetic contribution in the stabilization of the native state of RNAs , including complex formation with proteins and other macromolecules [56] . The dynamics of nucleic acids are also found to be strongly influenced by the motion of their ion atmospheres . Relative to other ionic species , Mg2+ can efficiently support a close assembly of negatively charged phosphates by mediating favorable interactions among them . Other earth alkali metals/divalent ions ( e . g . Ca2+ ) and even monovalent ions are also able to induce similar transitions , albeit at higher concentration . Our early SEC elution profiles for SAM-II show that the transition midpoint in presence of Potassium ( K+ ) alone occurs only at [K]1/2 ≈ 25 mM . The midpoint for Calcium ( Ca2+ ) is [Ca]1/2 ≈ 8 mM , compared to 6 mM for the Mg2+ ion [40] . This is a direct result of the larger charge/radius ratio of magnesium [40 , 57] . Thus , having these special characteristics , Mg2+ efficiently helps pre-organize the system and enables access to the partially collapsed states that are further stabilized by ligand binding . The general importance of Mg2+ for the stability of compact RNA structures supports a possibly universal role of conformational selection in ligand-binding RNAs , such as riboswitches , aptamers , and possibly protein-binding RNAs . A detailed thermodynamic understanding of the underlying landscape will indeed enable greater control of riboswitch regulation , highly sought after by researchers in synthetic biology who are currently employing riboswitches as ligand-dependent ‘knobs’ to control desired gene expression [58] .
Our all-atom structure-based model ( SBM ) has proven successful in describing the dynamics of numerous proteins and macromolecular complexes [59–63] . To elucidate RNA free energy landscapes under the influence of Mg2+ , models capable of quantitatively describing the ion atmosphere are needed , including ionic condensation around the negatively charged phosphate groups of RNA . Early studies have simply included electrostatic effects in SBM of RNA via repulsive Debye-Hückel interactions , thus treating all ions implicitly [29 , 30] . Recently , our group developed a more detailed model of RNA electrostatics and applied it within all-atom structure-based molecular dynamics simulations . Our model treats Mg2+ ions explicitly to account for ion-ion correlations neglected by mean-field theories [38] . The KCl buffer , which completes the experimental setup , is treated implicitly by a generalized Manning counter ion condensation model [38 , 64] , since mean-field theories correctly assess the charge densities of monovalent K+ and Cl- ions . Classical Manning counter-ion condensation theory was originally developed for understanding the low concentration limiting behavior of polyelectrolyte chains modeled with an infinite line of charge . Folded RNA , however , is not a line of charge . To account for the compact and irregular structures of RNAs and the effects of varying ion concentrations , we improve the Manning counter ion condensation model to handle electrostatic heterogeneity , making the condensed charge density a dynamical function of each phosphate coordinate . KCl screening is characterized by a Debye-Hückel potential . Removal of the continuum screening ions from the inaccessible volume of RNA is a substantial extension to Manning counter-ion condensation . The model has been tested against experimental measurements of excess Mg2+ associated with RNA , characterizing the Mg2+-RNA interaction free energy . This hybrid SBM has opened up new possibilities to study various structural and functional processes of RNA that are essentially controlled by ions [38] . In the present study we used this recently developed all-atom hybrid SBM to understand the conformational transition of SAM-II and the corresponding Mg2+ sensitivity . The energy function used in this model is given below , Φ=ΦSBM+ΦMg-Size+Φion-effect ( 1 ) where , ΦSBM is the all-atom SBM potential ensuring a global minimum in the landscape for the native state of RNA . The SBM potential is composed of two general types of interactions: ΦSBM=Φlocal+Φnon-local ( 2 ) where , Φlocal characterizes the local interactions that encode covalent bonds and torsional angles , maintaining the correct local geometry and chirality . Φnon-local comprises two non-local contributions: ( i ) an attractive term that is applied specifically to all tertiary interactions determined from the native structure , ( ii ) the general repulsive interactions , that describe the excluded volume by symmetric hard potentials ( to avoid any unwanted chain crossing ) . ΦMg-Size adds the excluded volume interactions involving the explicit Mg2+ ions , regulating RNA-Mg2+ and Mg2+-Mg2+ interactions . Φion-effect accounts for all interactions between charges in the system which consist of the fixed charge distribution of the RNA and the dynamic contribution from the ions . Mg2+ and phosphate charges interact via a Debye-Hückel potential with a screening term that depends , in turn , on the distribution of the monovalent ions . The monovalent ions , K+ and Cl- from the added salt , fall into two categories: screening ions and Manning condensed ions . The screening ion density is obtained using Debye-Hückel electrostatics . The density of the Manning-condensed ions is modeled as the sum of two normalized Gaussian distributions where the center of each Gaussian is located on the position of the negatively charged phosphate group . All the condensation variables along with the explicit Mg2+ and RNA coordinates are evolved with Langevin dynamics [38] . The mathematical formulations of all the terms and the related parameterizations are discussed in depth in section S4 in the S1 Text . The umbrella sampling method [65] was used to sample the conformational space of SAM-II riboswitch along the reaction coordinate , Q , which is the fraction of intra-molecular native contacts in the riboswitch . The Weighted Histogram Analysis Method [66] was then used to calculate the thermodynamic quantity , G ( Q ) . The detail is described in section S5 in the S1 Text . CEST data were collected using a pseudo-3D HSQC experiment with the B1 field offsets ( -600 to 600 Hz ) incremented in an interleaved manner with 3 references ( no CEST period ) [47] . A total of 1024x16 complex points were recorded [40] with 32 transients with a recovery delay of 1 . 5 s for a total experimental time of approximately 12 hr for each spin-lock field . A CEST saturation period of 100 ms was used for the base and 200 ms for ribose . The pulse program used was an adaptation of a previously published one without the need for selective pulses [47] . We used a two-state model to fit each of the three profiles of the selectively labeled carbon ( ribose C1’ and base C6 ) and quantitatively extracted the carbon chemical shift ( Δω ) , the exchange rate , and the population of the minor state based on the Bloch−McConnell 7x7 matrix [47] . The CEST data was plotted as I ( t ) /I ( 0 ) versus spin-lock offset ( Hz ) and was fit by numerically solving the matrix exponential for the CEST spin-lock period based on this 7x7 two-state Bloch-McConnell equation as described earlier [47 , 67] . In experiments , Fluorescence Resonance Energy Transfer ( FRET ) efficiency is the quantum yield of the energy transfer where a donor chromophore from its excited electronic state may transfer its energy to an acceptor chromophore through a non-radiative dipole-dipole coupling . The FRET efficiency varies with the separation between donor and acceptor fluorophores following the Fӧrster relation . For theoretical FRET predictions we use the Fӧrster relation where the value of Fӧrster radius is taken as 53Å [68] . We described it in detail in section S1 in the S1 Text . In SAXS experiments , the scattering intensity is measured from the electron density difference between the purified sample and that of the solvent/buffer . FoXs is a method that uses the Debye formula by which a theoretical scattering profile of a structure can be computed [48 , 49] . The detail is discussed in section S2 in the S1 Text .
|
The presence of positively charged metal ions is essential to maintain the structural fold and function of RNA . Among different metal ions , magnesium is particularly important for the stability of RNA because it can efficiently support a close assembly of negatively charged phosphate groups in an RNA fold . The SAM-II riboswitch is an example of a classical pseudoknot fold , which binds S-adenosyl methionine , stabilizing an alternate folded form to inhibit gene expression . In our early 13C- and 1H-chemical exchange saturation transfer ( CEST ) experiments , we found a conformational transition between a minor , partially closed and a major , open state conformation in the absence of ligand . Our CEST experiments at different magnesium concentrations now suggest that magnesium ions can induce a conformational pre-organization in the apo SAM-II riboswitch , which is expected to facilitate ligand binding . To understand the microscopic details of this magnesium-induced transition , we perform all-atom structure-based molecular simulations including electrostatics and explicit magnesium ions . Our free energy calculations reveal that the partially closed pre-organized state is further stabilized with increasing magnesium concentration . This is in excellent agreement with our 13C-CEST profile , SAXS , and size-exclusion chromatographic data , and with recent single molecule FRET experiments . Our results suggest that a sufficiently high concentration of magnesium is essential to pre-organize the apo SAM-II riboswitch .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"chemical",
"compounds",
"phosphates",
"fluorophotometry",
"elution",
"thermodynamics",
"research",
"and",
"analysis",
"methods",
"separation",
"processes",
"spectrum",
"analysis",
"techniques",
"rna",
"structure",
"fluorescence",
"resonance",
"energy",
"transfer",
"magnesium",
"chemistry",
"molecular",
"biology",
"spectrophotometry",
"free",
"energy",
"physics",
"biochemistry",
"rna",
"biochemical",
"simulations",
"rna",
"folding",
"nucleic",
"acids",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"chemical",
"elements",
"macromolecular",
"structure",
"analysis"
] |
2017
|
A magnesium-induced triplex pre-organizes the SAM-II riboswitch
|
The Metabolic Syndrome ( MetS ) is a complex , multifactorial disorder that develops slowly over time presenting itself with large differences among MetS patients . We applied a systems biology approach to describe and predict the onset and progressive development of MetS , in a study that combined in vivo and in silico models . A new data-driven , physiological model ( MINGLeD: Model INtegrating Glucose and Lipid Dynamics ) was developed , describing glucose , lipid and cholesterol metabolism . Since classic kinetic models cannot describe slowly progressing disorders , a simulation method ( ADAPT ) was used to describe longitudinal dynamics and to predict metabolic concentrations and fluxes . This approach yielded a novel model that can describe long-term MetS development and progression . This model was integrated with longitudinal in vivo data that was obtained from male APOE*3-Leiden . CETP mice fed a high-fat , high-cholesterol diet for three months and that developed MetS as reflected by classical symptoms including obesity and glucose intolerance . Two distinct subgroups were identified: those who developed dyslipidemia , and those who did not . The combination of MINGLeD with ADAPT could correctly predict both phenotypes , without making any prior assumptions about changes in kinetic rates or metabolic regulation . Modeling and flux trajectory analysis revealed that differences in liver fluxes and dietary cholesterol absorption could explain this occurrence of the two different phenotypes . In individual mice with dyslipidemia dietary cholesterol absorption and hepatic turnover of metabolites , including lipid fluxes , were higher compared to those without dyslipidemia . Predicted differences were also observed in gene expression data , and consistent with the emergence of insulin resistance and hepatic steatosis , two well-known MetS co-morbidities . Whereas MINGLeD specifically models the metabolic derangements underlying MetS , the simulation method ADAPT is generic and can be applied to other diseases where dynamic modeling and longitudinal data are available .
The simultaneous presentation of obesity , dyslipidemia , insulin resistance and hypertension is generally referred to as the Metabolic Syndrome ( MetS ) [1–5] . Together , these factors impose increased risk for the development of co-morbidities including cardiovascular disease , type 2 diabetes and non-alcoholic fatty liver disease [6 , 7] . MetS is considered to be the result of an imbalance in the mechanisms controlling dietary intake , energy expenditure , glucose handling and lipid homeostasis [8–10] . The high prevalence of obesity and MetS [11–15] , in combination with the heterogeneous presentation of MetS patients [16 , 17] , asks for the design of adequate treatment and prevention strategies . Clinical research on MetS is mostly cross-sectional in nature and tends to focus on either lipid or glucose metabolism , while both make up MetS . Knowledge on the interplay between these different metabolic components during the relatively slow progression into disease is therefore limited , but which a systems biology approach could provide . Modeling efforts in the past have shown that systems biology can be a powerful approach to gain both qualitative and quantitative insight in the inherently complex systems that drive the development of MetS . For example , Lu et al . [18] demonstrated how HDL raising modulators fail to reduce cardiovascular disease by elucidating the effects on the reverse cholesterol transport pathway . Likewise , Topp et al . [19] have identified different response pathways and physiological outcomes during prolonged hyperglycemia . However , these computational models were designed to model the healthy [18 , 20–29] or diseased state exclusively [19 , 30] . Furthermore , these models only describe short-term dynamics ( e . g . the postprandial response period ) and do not take into account long-term dynamics that may be expected to occur in progressive diseases such as MetS . Moreover , models that do explicitly study the gradual phenotype transition into a diseased state , have only considered either the lipid component [31] or the glucose component [32–35] of MetS . In addition , a long-term simulation method has been developed referred to as Analysis of Dynamic Adaptations in Parameter Trajectories ( ADAPT ) [36–38] . It infers time-varying parameters that gradually change over time , reflecting the slow change in regulation of metabolic processes during disease development . Therefore , ADAPT is a very powerful approach to study longitudinal development of diseases and therapeutic interventions and uses experimental data to infer adaptations in the system [36 , 37 , 39 , 40] . In previous studies , ADAPT has been applied to study hepatic steatosis [37 , 39] , and treatment of type 2 diabetes [40] , but has not yet been applied to study the full metabolic complexity of MetS . Therefore , we aimed to design a computational , data-driven approach to study the longitudinal and progressive dynamics of the majority of metabolic alterations of MetS , i . e . obesity , glucose intolerance , insulin resistance and dyslipidemia . We employed a systems biology methodology that integrates three main concepts to infer metabolic adaptations during MetS development: i ) the long-term simulation method ADAPT , combined with ii ) a newly developed in silico MetS model that describes the metabolic processes involved in whole-body carbohydrate and lipid metabolism , and integrated with iii ) time-series data obtained from an in vivo MetS model . The APOE*3-Leiden ( E3L ) . CETP mouse [41 , 42] is a mouse model for human MetS that develops diet-induced dyslipidemia and is prone to develop obesity and insulin resistance . It has been used in cross-sectional studies addressing different metabolic facets of MetS [41 , 43–51] . We made use of this model to study the development of diet-induced MetS in a longitudinal setting [48 , 52] by collecting measurements in the same animals within a three-month period and at multiple intermediate time points . Our in silico modeling quantitatively analyzes and integrates the experimental data and provides estimates of metabolite concentrations and fluxes that were experimentally not measured . With this modeling approach , we demonstrate when and how the onset and development of MetS occurs . Although we expected to find a homogeneous population , our modeling approach shows the emergence of different phenotypes of MetS . This heterogeneity is associated with differences in intestinal and hepatic metabolic fluxes .
Fig 1 presents the results of feeding LFD , HFD and HFD+C as average ( left ) and individually ( right ) for HFD+C feeding; the latter will be discussed in the next paragraph . All individual datasets are also available in S1 Data . As compared with LFD feeding , increased fat intake ( HFD ) resulted in increased weight gain ( Fig 1A ) reflected by a marked increase in fat mass ( Fig 1B ) with no changes in lean mass ( Fig 1C ) . HFD feeding also significantly increased fasting plasma glucose ( Fig 1D ) and insulin ( Fig 1E ) levels . Although HFD did not affect the plasma total cholesterol level ( TC; Fig 1G ) , a shift in the lipoprotein ratio was observed , reflected by increased plasma HDL-cholesterol ( HDL-C ) levels ( Fig 1H ) upon feeding a high-fat diet . As compared to the HFD alone , obesity development was slightly increased by cholesterol feeding ( Fig 1A and 1B ) , as monitored by increased fat mass at three months after starting the diet , even though the daily food intake did not increase ( data not shown ) . Although the additional dietary cholesterol did not affect plasma glucose and insulin levels ( Fig 1D and 1E ) , it did impair glucose tolerance in the oral glucose tolerance test ( see S1 Note–Fig 1 ) , indicating that dietary cholesterol increased insulin resistance . As compared to HFD , plasma triglyceride ( TG ) levels were significantly increased after three months of feeding HFD with cholesterol ( Fig 1F ) . Cholesterol feeding did not seem to have an effect on circulating TC levels until the second month of the dietary induction ( Fig 1G ) . Plasma HDL-C level ( Fig 1H ) , on the other hand , was increased after starting the diet ( HFD and HFD+C ) over the first two months . Male E3L . CETP mice developed HFD-induced obesity , glucose intolerance and dyslipidemia , mimicking the classical symptoms seen in human MetS . However , the high variability in both triglyceride and cholesterol levels complicated the interpretation of the dyslipidemic component . Therefore , we also presented the data of mice fed with the HFD+C individually in the right-hand side panels of Fig 1 . When inspecting the levels of plasma TG ( Fig 1F ) and plasma TC ( Fig 1G ) at the three months’ time point , the data reveal clear differences among the individual mice: while some develop dyslipidemia upon feeding HFD+C ( shown in red ) , others do not ( shown in grey ) , suggesting a bimodal distribution . With this insight , we divided the HFD+C cohort into two subgroups: the mode with high plasma TG and high TC levels after three months of HFD+C is referred to as the dyslipidemic Metabolic Syndrome phenotypes ( MetSDLP; n = 3 ) and the other mode as the non-dyslipidemic Metabolic Syndrome phenotypes ( MetSnon-DLP; n = 5 ) . This subdivision shows a consistent pattern in both TG and TC at the three months’ time point , but is already present earlier in time . In fact , the onset of dyslipidemia is already noticeable after two months of the diet , and progresses further towards the three months’ time period . Note that the subdivision in phenotype development is a result of the HFD+C diet; no baseline differences are observed ( t = 0 months in this data set ) , and these observations are limited to the HFD+C group and not present in HFD . This separation into two phenotypes is clearly present in the lipid trait , but it is also reflected in markers of carbohydrate metabolism . Although the individual mice cannot be separated based on the fasting plasma glucose ( Fig 1D ) data , the fasting plasma insulin ( Fig 1E ) indicates that the MetSDLP individuals are significantly more insulin resistant than the MetSnon-DLP individuals after at the three months’ time point . This trend is even more profound in the insulin dynamics in response to a glucose challenge test ( S1 Note-Fig 1 ) : MetSDLP individuals show a markedly higher insulin peak that also lasts longer than the MetSnon-DLP individuals . This indicates that dyslipidemia and glucose intolerance develop in parallel , but it is unclear how these are causally related . Next , we developed a novel in silico model that is tailored to describe data from longitudinal studies and simulates the metabolic system governing carbohydrate , lipid and cholesterol metabolism , providing predictions for unobserved fluxes and metabolites . The Model INtegrating Glucose and Lipid Dynamics ( MINGLeD ) is composed of a system of coupled , nonlinear ordinary differential equations ( ODEs ) . The steady state of the ODE model represents a snapshot of the metabolic state and describes the mass balance of the metabolite pools and flux rates for carbohydrate , lipid and cholesterol species in the plasma , liver , intestinal lumen and periphery ( Fig 2 ) . The model aims to describe the metabolic system from a whole-body perspective under healthy conditions as well as at different stages during MetS development . All simulation code and in silico data files are available on GitHub ( via https://github . com/yvonnerozendaal/MINGLeD ) . MINGLeD was integrated with the in vivo data whilst considering four subgroups: the LFD group , HFD group , MetSnon-DLP phenotypes and MetSDLP phenotypes . For each subgroup , MINGLeD was fitted to the data of each of the four time snapshots separately . The resulting sixteen models differ in the values for their estimated parameters . For each model the parameter estimation procedure was repeated with 500 different initial parameter sets using multi-start optimization . Fig 3 displays that MINGLeD accurately fits the metabolic snapshots over time for each of these groups . This shows that MINGLeD is capable of describing different metabolic phenotypes upon varying dietary intake and at different metabolic stages in time . Calibration of MINGLeD yielded separate models for each subgroup ( LFD , HFD , MetSnon-DLP and MetSDLP ) . However , this ignores the fact that the phenotypes represented by those four models are causally connected in time . MINGLeD describes metabolic fluxes and concentrations but does not explicitly include the multiple pathways that regulate and modulate metabolism over a three-month time period ( such as changes in gene expression and protein activity ) . Both limitations were overcome by combining MINGLeD with a dedicated approach for longitudinal modeling of biological systems: ‘Analysis of Dynamic Adaptations of Parameter Trajectories’ ( ADAPT ) [36–38] . ADAPT uses the experimental data to infer adaptations in the system , which is implemented by introducing time-varying model parameters . Model parameters are iteratively re-estimated over the time course of the simulation , yielding parameter trajectories that govern the time-dependent evolution of the modeled state variables and fluxes . Combining MINGLeD with ADAPT and the experimental data resulted in a dynamic , continuous model of MetS development and progression . ADAPT uses ensemble-based simulation to account for both methodological and experimental uncertainty resulting from biological variability and relatively low power in the HFD+C subgroups . By repeated Monte Carlo sampling of the in vivo data , sampled datasets are generated and for each sampled dataset a parameter trajectory was estimated . This was repeated resulting in a set of 1 , 000 simulated models yielding a database comprising of in silico populations of 1 , 000 virtual individuals for each subgroup . Each simulation describes the experimental data adequately , though with variation in parameter trajectories yielding differences in fluxes and concentrations , especially for model variables that are not experimentally observed . A more detailed explanation of the ADAPT methodology is given in the Online Methods . Fig 4 shows that the simulated trajectories for the plasma metabolites fit the in vivo data points accurately and provide a continuous description of the dynamics with which the system behaves over time , illustrating the different trajectories for the different subgroups . Note that ADAPT did not yield one single simulation but provided ensembles of state and flux trajectories . Each line in Fig 4 is one trajectory solution , where a darker color represents more overlap ( higher density ) of trajectory solutions in that region . These density plots cover the solution space for each of the modeled components . To aid visual analysis , the mean of the trajectory distributions for each of the subgroups is depicted in the panels to the far-right . These trajectories illustrate the clear distinction that can be made between the dyslipidemic and non-dyslipidemic MetS phenotypes . We showed that the in silico model accurately captures the trends as observed in the in vivo data . The ensembles of simulated concentration and flux trajectories provide insight in how the underlying metabolic network is affected upon emergence of the two different MetS phenotypes . Fig 5 displays the median with 10% range of the collection of trajectories for several metabolic fluxes . Clear differences between the dyslipidemic and non-dyslipidemic MetS phenotypes can be observed . A complete overview of all model state trajectories and metabolic flux trajectories is documented in S1 Fig . Despite equal intake of food and thus cholesterol , dietary cholesterol absorption from the intestinal lumen was markedly higher in the dyslipidemic compared to the non-dyslipidemic MetS group ( Fig 5A ) , which may have promoted the development of dyslipidemia . The modeling also predicted that within the dyslipidemic phenotype , rates of lipid shuttling towards ( V ) LDL-TG uptake ( Fig 5B; see also S1K–S1L Fig ) were increased , as higher rates of hepatic fatty acid uptake ( Fig 5C ) and lipogenesis ( Fig 5E ) were observed accompanied by a lower rate of hepatic β-oxidation ( Fig 5F ) . We observed similar changes if we analyze the gene expression data ( S1 Note-Table 3 ) from the same animals: we could qualitatively relate our flux predictions with the gene expression profiles of the fully developed MetS phenotype after three months of dietary induction . We found , as compared to LFD , an upregulation of hepatic fatty acid uptake in the HFD , HFD+C ( MetSnon-DLP ) and HFD+C ( MetSDLP ) groups ( Fig 5C ) . Fig 5D shows that hepatic bile acid synthesis is predicted to be higher in the MetSDLP phenotype . This aspect could also be observed in the gene expression data: the expression of bile acid synthesis genes nuclear farnesoid X receptor ( FXR ) and Cholesterol 7 alpha-hydroxylase ( Cyp7a1 ) was largely upregulated , as compared to HFD alone . Hepatic de novo lipogenesis ( Fig 5E; DNL ) is predicted to be higher in the dyslipidemic phenotype , for which data at the 3 months’ time point was available and was used in the model fitting . Genes related to DNL , including Fatty Acid Synthase ( FASN ) , Acetyl-CoA carboxylase 2 ( ACC2 ) , acyl-CoA:diacylglycerol acyltransferase 2 ( DGAT2 ) and sterol regulatory element binding protein 1c ( SREBP1c ) were all largely downregulated probably as a compensatory mechanism . In addition , the hepatic β-oxidation ( Fig 5F ) is predicted to be lower in the dyslipidemic group . This could be linked to a compensatory upregulation of genes related to fatty acid β-oxidation such as Peroxisome proliferator-activated receptor α ( PPARα ) , Peroxisomal acyl-coenzyme A oxidase 1 ( ACOX1 ) and Carnitine palmitoyltransferase 1a ( CPT1a ) . Collectively , these predictions suggest more lipid accumulation in the liver of the dyslipidemic mice . MINGLeD with ADAPT predicted the trajectories of the liver lipid profiles and calculated the hepatic lipid pool sizes over the entire time course of three months . Next , we measured lipid turnover and the lipid content in livers from individual mice after three months of dietary induction to verify these in silico predictions at the last time point . Indeed , addition of cholesterol did lead to an increase in lipid turnover and the hepatic lipid pool sizes as compared to the high-fat diet without cholesterol , both in the dyslipidemic and non-dyslipidemic phenotype ( Fig 6 ) . Moreover , this accumulation of lipids in the liver , in particular cholesterol components , was more profound in MetSDLP mice . Flux trajectory analysis revealed substantial differences in hepatic fluxes between the non-dyslipidemic and the dyslipidemic MetS phenotypes . In terms of plasma metabolite levels ( see Fig 4A–4E ) , MetSnon-DLP mice closely resemble the mouse population fed the HFD without cholesterol , and intriguingly , appears not to be affected by the additional dietary cholesterol load . MINGLeD with ADAPT predicted that this may be due to reduced dietary cholesterol absorption ( Fig 5A ) in the non-dyslipidemic phenotype . Furthermore , the predicted lipid pool sizes were in agreement with liver histology data ( S1 Note-Fig 2 ) , which showed the establishment of microvesicular steatosis upon HFD feeding . In contrast , steatosis was exacerbated in MetSDLP mice revealed by a more severe type of macrovesicular steatosis .
The clinical presentation of MetS in humans is highly heterogeneous and spans over decades . Male E3L . CETP mice fed a high-fat diet supplemented with cholesterol develop MetS within a time scale of several months . Although all of these animals have the same genetic background , received the same diet and were kept and monitored in a controlled , standardized environment , this in vivo model did show heterogeneity in phenotypic presentation . In addition , the manifestation of the full repertoire of metabolic alterations associated with MetS makes this a useful in vivo model , whereas other animal models only describe one or partial metabolic aspects of MetS [53–59] . Using a traditional statistical approach , both this heterogeneity and limited datasets comprising of low number of animals are problematic . Moreover , the time-dependency of the data–i . e . individual data of consecutive points in time are interrelated and therefore not independent samples–would further complicate analysis . Our computational modeling approach tackles these problems by combining MINGLeD with ADAPT . Contrary to other computational models , MINGLeD integrates both glucose and lipid species at a whole-body level . Both carbohydrates and fats are of importance in MetS and MINGLeD allows for simultaneous description of these key components in terms of both metabolite pool sizes ( concentrations ) and metabolic fluxes . Complexity and detail in model equations was considered in close relation to what is experimentally feasible to measure throughout long-term MetS development . Therefore , MINGLeD’s data-driven , physiological design allows for describing both flux and concentration data on a whole-body level . MINGLeD per se can be simulated to describe metabolic snapshots , whereas the long-term dynamics are captured by using MINGLeD in conjunction with ADAPT . ADAPT has been designed to work with this kind of data and makes use of the time-dependent observations and simultaneously assesses uncertainty based on the variability in the data . The strength of the ADAPT methodology in dealing with heterogeneity became evident with the identification of two distinctly different phenotypes despite the limited number of animals that were studied . These two phenotypes mainly differ in terms of dyslipidemia , and flux trajectory analysis pinpointed differences in: 1 ) hepatic turnover of metabolites , including lipid fluxes and 2 ) the intestinal cholesterol absorption . This appears to mimic the observation in humans showing that levels of cholesterol absorption efficiency can vary greatly among individuals [60 , 61] . These model predictions are open for further experimental validation . The methodology of integrating in vivo and in silico information allows to combine pre-existing knowledge with experimental quantitative data and can therefore be applied to study other multifactorial , progressive diseases where longitudinal data are available . A future application of the model is to quantify energy intake and energy expenditure and analyze the energy balance over time for development of obesity and MetS . In conclusion , we combined data from animal experiments with a computer model and computer simulations to study the development of MetS . The new model predicted which changes in the underlying metabolic processes could explain the MetS symptoms . Two different subgroups were identified: those with high cholesterol and high triglycerides , and those without . The computer model found that in those who develop lipid abnormalities , both dietary cholesterol absorption and hepatic liver fluxes were higher .
All animal experiments were performed in accordance with the regulations of Dutch law on animal welfare , and the Animal Ethics Committee of the Leiden University Medical Center , Leiden , The Netherlands . Animals were sacrificed by CO2 inhalation . The first step of our systems approach involves the gathering of in vivo data at different stages during the development of the Metabolic Syndrome . To this end we use male APOE*3-Leiden ( E3L ) . CETP transgenic mice as diet-induced in vivo model to study the metabolic adaptations that occur over time . Male E3L . CETP transgenic mice were housed under standard conditions with a 12 h light/dark cycle ( 7AM-7PM ) , housed with 1–2 animals per conventional cage with free access to chow diet and water , unless indicated otherwise . At the age of 11 weeks , randomized according to body weight and plasma lipids ( total cholesterol and triglycerides ) and glucose , mice were divided into three groups: mice were fed either a low-fat diet ( LFD; n = 8 ) , high-fat diet ( HFD; n = 12 ) or a HFD with additional cholesterol ( HFD+C; 0 . 25% , w/w; Sigma ) ( HFD+C; n = 8 ) for three months . The LFD has a 20% energy content derived from lard and contains 3 . 8 kcal/g diet; the HFDs have a 60% energy content derived from lard and contain 5 . 2 kcal/g diet ( OpenSource Diets , Research Diets , Inc . New Brunswick , USA ) . The specific composition of each diet is listed in S1 Note-Table 1 . During the study , body weight and food intake were measured weekly , body composition ( lean and fat mass ) every other week . Blood samples were taken monthly and analyzed for glucose , insulin , free fatty acids , total cholesterol , HDL-cholesterol and triglycerides . At the end of the three months dietary induction experiment , animals were sacrificed . Livers were isolated for measuring lipid metabolites and gene expression , and the hepatic de novo lipogenesis was determined by using an isotope tracer . For further details on the experimental setup the reader is referred to S1 Note . All individual data is available as S1 Data . The second step of our systems approach involves the development of a dedicated in silico model describing MetS onset and progression . To this end we have developed a novel dynamic , computational model called Model INtegrating Glucose and Lipid Dynamics ( MINGLeD ) . This model is tailored to describe data from longitudinal studies and involves the metabolic system governing carbohydrate , lipid and cholesterol metabolism . It is composed of a system of coupled , nonlinear ordinary differential equations ( ODEs ) . The steady state of the ODE model represents a snapshot of the metabolic state and describes the mass balance of the metabolite pools and flux rates for carbohydrate , lipid and cholesterol species in the plasma , liver , intestinal lumen and periphery . This in silico model aims to describe the metabolic system in a whole-body perspective under healthy conditions as well as at different stages during Metabolic Syndrome development . The steady state of MINGLeD describes the average behavior of the metabolic system over the time course of one day . Therefore no specific postprandial or fasting periods have been included explicitly . The metabolic pathways that are important in describing the metabolic system are presented in Fig 2 . The exact block diagram of the computational model is schematically represented in S2 Note-Fig 1 . The system of ODEs and fluxes equations are derived in S2 Note and listed in S2 Note-Tables 1–2 . MINGLeD describes metabolite pools originating from carbohydrate substrates ( glucose , glucose-6-phosphate , acetyl coenzyme a ) , lipid species ( free fatty acids , triglycerides , various lipoproteins ) and cholesterol ( free cholesterol , cholesteryl esters and bile acids ) . The metabolic pathways that define the interactions between these metabolites include the uptake of dietary macronutrients , glycolysis , hepatic gluconeogenesis , lipoprotein assembly and ( remnant ) uptake , cholesteryl ester transfer between lipoproteins ( upon action of the cholesterol ester transfer protein; CETP ) , trans-intestinal cholesterol excretion ( TICE ) , exchange between free cholesterol and cholesteryl esters ( via ACAT and CEH ) in hepatic tissue , hepatic fatty acid uptake , peripheral lipolysis , β-oxidation , de novo lipogenesis ( DNL ) and cholesterol biosynthesis . Bile acid synthesis , biliary bile acid and cholesterol excretion , enterohepatic reuptake , fecal excretion of bile acids and cholesterol . Finally , metabolism of acetyl coenzyme A in hepatic and peripheral tissues has been included . To validate the proposed structure of the in silico model , we calibrate MINGLeD to the in vivo data of the different metabolic snapshots . The experimentally measured metabolites are closely related to many of the variables in the computational model , such that an identifiable model is achieved of which the model parameters can be determined using parameter estimation . For each of the diet cohorts , MINGLeD is fitted separately to the phenotype snapshot determined at each month during the dietary induction period . For each snapshot , we compute the average and standard deviation of the measured data at this time point and use this data to fit the model to using maximum likelihood estimation . S3 Note and S3 Note-Table 1 describe how we relate the experimentally observed data to the specific model outputs and equations using the weighted sum of squares as error measure to be minimized . A Monte Carlo approach is employed to account for methodological and experimental uncertainties . The optimization procedure is repeated 500 times using a widely dispersed range of initial parameter values ( 10−1–101 ) to accommodate multi-start optimization . The implementation details can be found in S2 Note . Once optimized , MINGLeD predicts unobserved species and fluxes and is used to make predictions on the underlying differences between phenotype snapshots . The next step in our systems approach is to couple the phenotype snapshots in time . Hereto we make use of the computational technique entitled ‘Analysis of Dynamic Adaptations of Parameter Trajectories’ ( ADAPT ) [36–38] . By employing this concept of the time-dependent evolution of model parameters , dynamic disease trajectories are obtained from which the onset and progression of MetS symptoms and co-morbidities can be studied . The progression of these adaptations is predicted by identifying necessary dynamic changes in the model parameters to describe the transition between experimental data observed at different points in time during the dietary induction . Since it is a priori unknown which model parameters change with time , it is not possible to perform a dynamic simulation of the entire time span in one go . Therefore we discretize the time span into 90 segments , each representing one day . ADAPT interpolates between the individual snapshots in time ( at which experimental data was obtained ) and simulates every day in between these time points . To facilitate this , some pre-processing steps are required . Since the quantitative experimental data is discrete and only available at four points in time , the data is interpolated using cubic smoothing splines to obtain continuous dynamic descriptions of the experimentally observed metabolite pools and fluxes . To account for experimental and biological uncertainties , a collection of 1 , 000 splines is calculated using a Monte Carlo approach: different random samples of the experimental data are generated assuming Gaussian distributions with means and standard deviations of the data . Subsequently , for each generated sample , a cubic smoothing spline is calculated . This bootstrapping approach yields samples of data replicates which will subsequently be utilized in parameter estimation [62] . By combining bootstrapping of data , sampling of parameters and a robust optimization of model simulations , ADAPT provides feedback about uncertainty in model predictions accounting for uncertainty in both experimental measurements and fitting procedures . The system is first simulated for the phenotype prior to the dietary induction by optimization to the t = 0 data . For each step in time , the system is simulated using the final values of the model states of the previous time step as initial conditions . Since we consider the parameters to be time-varying , the model parameters are re-optimized for each time step by minimizing the difference between the ( sampled and interpolated ) experimental data and the corresponding model outputs . The estimated parameter set from the previous time point is used as initial set for the optimization procedure . It is assumed that the induced adaptations are minimal and proceed progressively in time . Therefore , highly fluctuating parameter trajectories are considered to be non-physiological . To prevent the occurrence of such behavior , the parameter estimation protocol is extended to prevent unnecessary change of parameters and to identify minimal parameter adaptations that are required to describe phenotype transition . The cost function is extended with a regularization term ( see Eq 2 in S3 Note ) , given by the sum of squared derivatives of the normalized parameter values . Hence , changing a parameter is costly , and will therefore be avoided if this is not required to describe the experimental data . The constant λ that determines the strength of this regularization term , should be chosen carefully such that the data fitting is biased as little as possible . If too large a value for λ is chosen , the regularization term becomes dominant and the model will not describe the experimental data accurately anymore . Since a small λ is already sufficient to minimize parameter changes and fluctuations , whilst still describing the experimental data accurately , λ was set to 0 . 1 in this study . All simulation code and in silico data files are available on GitHub ( via https://github . com/yvonnerozendaal/MINGLeD ) . ADAPT yields a collection of parameter sets that describe the dynamics of the onset and development of MetS over time . The obtained trajectories for the model parameters , but also for the fluxes and pool sizes provide insight in the affected underlying biological system . This provides information about the adaptations that have taken place during the dietary induction , and these model-based predictions are compared to the gene expression data that is measured at the end of the dietary induction study .
|
Obesity is becoming a growing health problem , with the risk of running into a disease state called the Metabolic Syndrome ( MetS ) . It is difficult to study MetS in humans as it develops over a long period of time while many processes and organs are involved . Moreover , each patient may present itself with different characteristics . We combined data from animal experiments with a computer model and computer simulations to study the development of MetS . The new model correctly predicts which changes in the underlying metabolic processes could explain the MetS symptoms . Unexpectedly , two different subgroups were identified: those with high cholesterol and high triglycerides , and those without . The computer model found that in those who develop lipid abnormalities , both dietary cholesterol absorption and hepatic liver fluxes were higher .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"cholesterol",
"biochemistry",
"carbohydrate",
"metabolism",
"lipids",
"medicine",
"and",
"health",
"sciences",
"dyslipidemia",
"nutrition",
"lipid",
"metabolism",
"biology",
"and",
"life",
"sciences",
"diet",
"metabolism",
"metabolic",
"disorders",
"glucose",
"metabolism"
] |
2018
|
In vivo and in silico dynamics of the development of Metabolic Syndrome
|
The crystal structure of the TLR4-MD-2-LPS complex responsible for triggering powerful pro-inflammatory cytokine responses has recently become available . Central to cell surface complex formation is binding of lipopolysaccharide ( LPS ) to soluble MD-2 . We have previously shown , in biologically based experiments , that a generation 3 . 5 PAMAM dendrimer with 64 peripheral carboxylic acid groups acts as an antagonist of pro-inflammatory cytokine production after surface modification with 8 glucosamine molecules . We have also shown using molecular modelling approaches that this partially glycosylated dendrimer has the flexibility , cluster density , surface electrostatic charge , and hydrophilicity to make it a therapeutically useful antagonist of complex formation . These studies enabled the computational study of the interactions of the unmodified dendrimer , glucosamine , and of the partially glycosylated dendrimer with TLR4 and MD-2 using molecular docking and molecular dynamics techniques . They demonstrate that dendrimer glucosamine forms co-operative electrostatic interactions with residues lining the entrance to MD-2's hydrophobic pocket . Crucially , dendrimer glucosamine interferes with the electrostatic binding of: ( i ) the 4′phosphate on the di-glucosamine of LPS to Ser118 on MD-2; ( ii ) LPS to Lys91 on MD-2; ( iii ) the subsequent binding of TLR4 to Tyr102 on MD-2 . This is followed by additional co-operative interactions between several of the dendrimer glucosamine's carboxylic acid branches and MD-2 . Collectively , these interactions block the entry of the lipid chains of LPS into MD-2's hydrophobic pocket , and also prevent TLR4-MD-2-LPS complex formation . Our studies have therefore defined the first nonlipid-based synthetic MD-2 antagonist using both animal model-based studies of pro-inflammatory cytokine responses and molecular modelling studies of a whole dendrimer with its target protein . Using this approach , it should now be possible to computationally design additional macromolecular dendrimer based antagonists for other Toll Like Receptors . They could be useful for treating a spectrum of infectious , inflammatory and malignant diseases .
Dendrimers are a class of spherical macromolecules that can be distinguished from conventional linear polymers by their highly branched and symmetrical architecture . Polyamidoamine ( PAMAM ) dendrimers are , by far , the best studied of the commercialised and divergently synthesised dendrimers . Typically , these dendrimers are available in whole generations ( amine terminated ) and half-generations ( carboxylic acid terminated ) that are representative of both their size ( i . e . , diameter in angstroms ) and molecular weight [1]–[3] . They can be made by controlled sequential processes to give well defined chemical structures . Even at low concentrations , the peripheral amine groups of cationic dendrimers damage cell membranes and lead to cell toxicity [4] . In contrast , anionic dendrimers have:- ( i ) physico-chemical properties that are similar to those of conventional small molecule drugs; ( ii ) can be modified to exist as zwitterions at physiological pH; ( iii ) have considerable buffering capacity that makes them physico-chemically “similar” to albumin , and therefore biocompatible . However , unlike proteins , they:- ( a ) do not undergo proteolytic degradation in plasma; ( b ) are not immunogenic or otherwise toxic even after repeated administration by various routes to animals; ( c ) can be optimized for their circulation time; ( d ) show preferential accumulation in tissues containing inflammatory cells compared to healthy tissue at a ratio of 50∶1 [4] . In addition , the National Cancer Institute's Nanotechnology Characterisation Laboratory recently undertook detailed chemical and toxicological characterization of anionic PAMAM dendrimers and found them to be both stable and biocompatible [5] . Taken together , these observations suggest that anionic dendrimer based drugs could become a new and safe class of “synthetic baby-bio” ( SBB ) drugs . In biologically based experiments , we have already shown that a generation ( G ) 3 . 5 PAMAM dendrimer that was partially modified with an average of 8 surface glucosamine molecules inhibited TLR4-MD-2-LPS pro-inflammatory cytokine mediated inflammation in primary human monocytes , dendritic cells , and in a clinically validated rabbit model of tissue scaring [6] . Molecular modeling studies suggested , and experimental studies confirmed , that the surface loading of a G3 . 5 PAMAM dendrimer ( with an average of 64 peripheral carboxylic acid groups ) could not be increased beyond an absolute maximum cluster density of 12 evenly spaced surface glucosamine molecules using the divergent synthesis approach [7] . Frontier molecular orbital theory ( FMOT ) and molecular dynamics simulations also showed that the optimum surface loading and distribution of the zero length amide bond conjugated glucosamine molecules was determined by both electronic effects and the different dynamic conformations adopted by the modified dendrimer during the incremental addition of glucosamine [7] , [8] . Importantly , the structural features and the dynamic behavior of this partially glycosylated dendrimer showed that its flexibility and polarity changed with the incremental addition of glucosamine molecules . Notably , these surface glucosamine molecules remained available for interaction with the biological target . Innate immunity provides immediate and efficient defence against microbial infection and tissue injury by promoting pro-inflammatory cytokine responses that induce adaptive immune responses . Toll-like receptors ( TLRs ) , a family of type I transmembrane glycoproteins , are central to these vertebrate innate immune responses because they recognize a broad range of soluble microbial stimuli on pathogens [9] . They are therefore called pattern recognition receptors . Their extracellular segments consist of leucine-rich repeats ( LRR ) with horseshoe-like shapes [10] . The binding of soluble agonist ligands leads to protein conformational change , receptor complex rearrangement , recruitment of specific adaptor proteins to the intracellular domain , and the initiation of signalling cascades . Lipopolysaccharide ( LPS – variable MWt>10 kDa ) is the outer membrane glycolipid of Gram-negative bacteria that induces this innate immune response [11] . It is composed of:- ( a ) the hydrophilic polysaccharide core; ( b ) the solvent exposed hydrophobic lipid A component; ( c ) the O-antigen . Only the lipid A is required to induce pro-inflammatory cytokine responses . It is composed of a diphosphorylated β-1 , 6-linked D-glucosamine disaccharide linked via amide or ester bonds to 3-hydroxy fatty acids further substituted by nonhydroxylated 12–14 carbon fatty acid chains [12] . The cell surface interaction between LPS , TLR4 and MD-2 protein is central to the initiation of pro-inflammatory cytokine mediated responses . The transport protein CD14 first collects and delivers soluble LPS to soluble , circulating and monomeric MD-2 [13] , [14] . The two phosphorylated glucosamine residues of lipid A ( MWt∼2 kDa ) bind via electrostatic interactions to the charged entrance ( Ser118 {for the 4′ phosphate} and Lys122 {for the 1′ phosphate} , and to Arg90 and Lys91 ) of human MD-2's hydrophobic pocket . The diglucosamine and phosphate groups remain in solution and outside the hydrophobic pocket; these residues serve to anchor LPS to the entrance of MD-2's hydrophobic pocket . This is followed by the lipid chains of LPS becoming buried in MD-2's hydrophobic pocket [15] . Subsequent formation of the human TLR4-MD-2-LPS complex requires:- ( a ) hydrophobic interactions between Met85 , Leu87 , Ile124 and Phe126 on human MD-2 with Phe436 , Phe436 , Phe440 and Phe444 respectively on human TLR4; ( b ) hydrogen bond interactions between Arg90 and Gly123 on human MD-2 with Glu439 and Ser416 respectively on TLR4 [16] . The complex formed undergoes conformational changes that lead to receptor complex dimerization , triggering of intracellular signaling events , and the initiation of pro-inflammatory cytokine production [17] , [18] . Low level stimulation of pro-inflammatory cytokine production is physiologically beneficial for dealing with infections because it enables the activation of co-stimulatory molecules and the generation of adaptive immune responses . However , excess pro-inflammatory cytokine production can become pathological with serious adverse effects on the host which include septic shock and death [19] . Previous glycodendrimer based studies have used fully glycosylated molecules; i . e . , all of the peripheral groups of the dendrimer linked to saccharides using a spacer arm derived amide bond [20] , [21] . In contrast , our studies are the first to use a partially glycosylated dendrimer; i . e . , a small number of glucosamine molecules linked to the dendrimer's surface using a zero length amide bond . In order to better understand the molecular mechanism responsible for the ability of partially glycosylated dendrimers to block pro-inflammatory cytokine production , we studied their interaction with the LPS recognition system using molecular modelling techniques . We started from the premise that their inherent flexibility and , potentially , their size ( MWt = 13 . 6 kDa for a G3 . 5 PAMAM dendrimer ) were important molecular determinants of their biological properties [6] . However , the major limitation for structural biology studies was the absence of structural data . Although molecular dynamics simulation based study of dendrimer-biomacromolecules interactions is possible [22] , [23] , we did not have any information about the initial complex or a starting point for the simulation . We therefore used molecular docking as a tool to investigate the interactions between the partially glycosylated dendrimer and the LPS recognition system . To our knowledge , no previous molecular modelling study has reported the docking of a whole dendrimer with a protein target . Our studies therefore provide new insights that should enable the design of new macromolecules as novel antagonists of biologically important Toll like Receptor-ligand interactions [24] .
We first chose and validated molecular docking software packages that used protein as the target and synthetic macromolecules as the ligand . These are described in the Methods section and results shown in the Supplementary Figures S1 , S2 , S3 , S4 , S5 , S6 , and S7 . Amongst docking softwares that deal with ligands of increasingly larger size , we found that Patchdock [25] , Hex [26] and Glue [27] could be used . These packages were validated by reproducing the correct orientations of the ligands in:- ( i ) the crystal structure of the mouse TLR4-MD-2 complex ( PDB entry: 2z64 ) ; ( ii ) the crystal structure of lipid A complexed with human MD-2 ( PDB entry: 2z59 ) . We then studied the possible binding modes between the LPS recognition system and structures of both the biologically inactive unmodified dendrimer and the biologically active partially glycosylated dendrimer . By subtracting the results for the unmodified dendrimer from those for the partially glycosylated dendrimer , it was possible to determine the contribution of the surface glucosamine molecules . The first studies involved the docking of the glucosamine molecules ( as the moiety most likely to contribute to the biological activity of the partially glycosylated dendrimer ) with the human TLR4-MD-2 complex ( PDB entry: 3FXI ) using Glue . As the binding site for glucosamine is not known , the whole TLR-MD-2 complex was used as a target by employing a set of overlapping boxes to increase the resolution of the docking solution . The results of this series of docking studies were overlapped and the solutions ranked according to the interaction energy observed ( Figure 1 ) . The multiple binding sites of glucosamine molecules and their distribution on the surface of the protein complex meant that they were not suitable for the use of a “grow-out” strategy to determine the interactions between these molecules [23] . Nevertheless , it was notable that the three sites with a high number of poses and with high favourable interaction energies of the glucosamines with the TLR4-MD-2 complex were found to be along its beta-6 strand , the beta-7 strand , and the loop between amino acids Ile94 and Phe104 . They line the opening of MD-2's hydrophobic pocket ( Figure 1 ) . These results led us to focus our docking studies with the whole dendrimer on MD-2 as the primary target . MD-2 is a 160 amino acid glycoprotein with a MWt of 20 kDa [14] . It represents a class of MD-2–related lipid recognition ( ML ) proteins . It is folded into a single domain that consists of two ß sheets in the immunoglobulin fold . One sheet consists of three antiparallel ß strands , and the other sheet consists of six antiparallel strands . Between these sheets is a β cup topology lipid binding pocket with a volume of 1 , 710 Å3 and approximate dimensions of 15 Å by 8 Å by 10 Å . The ß6 and ß7 strands line the entrance to the hydrophobic pocket . Its shape suggests that it has evolved to accommodate large and structurally diverse ligands . Monomeric MD-2 binds to LPS and then forms a stable complex with TLR4 on the cell surface . Initially , the interaction between these dendrimers and MD-2 ( i . e . , shape complementarity ) was studied using the protein-protein interaction softwares Patchdock and Hex . They can carry out rigid docking only . In the absence of information about the dendrimer's bioactive conformation , we had to use different conformations of the dendrimers for these rigid docking studies . We had already shown that the partially glycosylated dendrimer was very flexible by molecular dynamics simulations of fully solvated dendrimers [7] , [8] . Initial 3D structures for the simulation were generated using our “sequence to conformation” method [7] , [8] . Twenty representative conformations were then selected from these molecular dynamics trajectories . Both the unmodified dendrimer and the dendrimer with 8 surface glucosamines were docked as ligands using Patchdock and Hex . The unmodified dendrimer was used as a negative control because it did not alter the biological activity of the TLR4-MD-2-LPS cell surface receptor complex . The target used was the crystal structure of human MD-2 ( PBD entry: 2e59 ) . For each ligand , the 20 lowest energy solutions were saved ( a total of 400 solutions per dendrimer ) and a rebol script implemented in the Vega ZZ interface used to process the data . The results were analysed in terms of the total number of interactions between atoms in the dendrimer and the residues of MD-2 for both the biologically inactive unmodified dendrimer , and the biologically active partially glycosylated dendrimer . Although both dendrimers revealed similar interaction profiles , the partially glycosylated dendrimer was found to have better shape complementarity , and it formed more interactions with MD-2 . These results showed that the surface conjugated glucosamines were important for the dendrimer's interaction with MD-2 . To investigate if there was specificity in the interactions observed using Patchdock , the solvent accessible surface area ( SASA ) of each residue of MD-2 was determined ( Figure 2A ) , and plotted alongside the number of interactions for the partially glycosylated dendrimer ( Figure 2B ) . We found that the overall negative charge of the unmodified dendrimer's surface [7] , [8] did not favour interactions with the negatively charged surface residues Glu92 , Asp99 and Asp101 that line MD-2's hydrophobic pocket . This is the most likely explanation for the lack of biological activity of the unmodified dendrimer . Comparing the interaction profiles for unmodified dendrimers with those for partially glycosylated dendrimers also revealed that the most exposed residues were not always the ones with the highest number of interactions with MD-2 . This indicated that there was selectivity in the interactions between the partially glycosylated dendrimer and MD-2 . The same conformations of both dendrimers were then submitted for docking studies with the Hex shape and electrostatics protocol ( Figure 2C ) . Although there was some similarity between the interaction profiles for Patchdock and Hex , a larger number of interactions were detected by Hex for both dendrimer types and the opening of MD-2's pocket . Comparing the results from the two software packages emphasised the importance of electrostatic interactions between the partially glycosylated dendrimer and MD-2 . A subtraction of the profile for the partially glycosylated dendrimer from the unmodified dendrimer showed significant differences ( Figure 3 ) . Most of the interactions of the unmodified dendrimer were with residues located along the sides and back of MD-2's pocket . In contrast , the partially glycosylated dendrimer had multiple interactions with the entrance to the opening of MD-2's pocket . These results suggested that the biologically important interactions of the partially glycosylated dendrimer were occurring with residues that lined the entrance to MD-2's pocket . These encouraging results from Hex were then normalised with the energy value determined by multiplying the number of interactions by the corresponding interaction energy for each solution . When interaction energies were also taken into account , the differences between the unmodified dendrimer and the partially glycosylated dendrimer became even more marked for the residues lining the entrance to MD-2's hydrophobic pocket ( Figure 4 ) . The surface residues lining the entrance of human MD-2's pocket that have been shown to have a key role in the electrostatic binding of LPS are Arg90 , Lys91 , Ser118 and Lys122 ( Figure 5A ) [15] , [17] . The biologically active partially glycosylated dendrimer showed the largest number and the strongest interactions with several of the residues lining the entrance to MD-2's pocket . Several of these residues are also important for the binding of LPS to MD-2 . The residues with the highest normalized interaction values were Lys91 , Tyr102 , Arg106 , Asn114 and Ser118 . Several other residues , with lower interaction values , also contributed significantly to the co-operative binding of the partially glycosylated dendrimer to human MD-2; they were Arg96 , Ser98 , Lys109 , Thr112 and Thr116 ( Figures 5B & C ) . To assess the dynamic behaviour of these interactions , and to determine whether using docking software designed for studying protein-protein interactions was affecting the results , a 4 . 8 ns molecular dynamics simulation of the partially glycosylated dendrimer complexed with MD-2 was performed in explicit solvent . The starting conformation used was one of the most stable docking solutions obtained with Hex . In this docking solution structure , the partially glycosylated dendrimer was located directly in front of MD-2's pocket with three of the glucosamine molecules in close proximity at 0 ns ( Figure 6 ) . As the simulation progressed , the partially glycosylated dendrimer moved away from MD-2 in order to rearrange its peripheral branches . This changed the number of glucosamine molecules facing MD-2's pocket . From 2 . 4 ns until the end of the simulation ( i . e . , 4 . 8 ns ) , the partially glycosylated dendrimer underwent further conformational changes such that 4 glucosamine molecules were consistently in close contact ( i . e . , 1 . 3 Å ) with the entrance of MD-2's pocket ( Figure 6 ) . In addition , the entrance to this pocket was occluded by the partially glycosylated dendrimer during the entire 4 ns simulation . This is shown in Figure 7 by the semi-transparent pink colour which represents the partially glycosylated dendrimer's structure during its interaction with the entrance of MD-2's hydrophobic pocket . The affinity of the partially glycosylated dendrimer for human MD-2 was demonstrated by the increased number of close contacts ( i . e . , 1 . 3 Å ) between these two molecules ( Figure 8 ) that involved both the glucosamine molecules and several of the dendrimer's peripheral carboxylic acid branches ( Figure 9 ) . These electrostatic interactions occluded the entrance to human MD-2's hydrophobic pocket and blocked access of the lipid chains of LPS . When taken with our previous observations that partially glycosylated dendrimers are both flexible and dynamic , this meant that conformational changes could induce shape complementarity [7] , [8] . This enabled the dendrimer's surface glucosamine molecules to block the entrance of human MD-2's pocket . Additional co-operative electrostatic interactions with some of the dendrimer's free carboxylic acid branches follow . Collectively , they block the entry of the lipid chains of LPS into human MD-2's pocket , and also prevent TLR4-MD-2-LPS cell surface complex formation . This is shown schematically in Figure 10 . The biologically important outcome is that the pro-inflammatory cytokine cascade is not initiated . Lipid A is the primary immuno-stimulatory core of LPS . Although diverse across bacterial species , its di-glucosamine portion is universally conserved . Variants of LPS can be discriminated by the TLR4-MD-2 complex as being either endotoxic ( i . e . , agonist ) or anti-endotoxic ( i . e . , antagonist ) or partial antagonist/partial agonist . For example , lipid A ( with six acyl chains ) from Escherichia coli is a potent agonist of cytokine production in human and mouse macrophages [28] . The lipid chains form co-operative interactions with hydrophobic residues in MD-2's pocket . However , its precursor , lipid IVa , which has only four acyl chains , is an antagonist in human macrophages but an agonist in mouse macrophages . These lipid chain based differences have been used to define the central role of the di-glucosamine moiety with respect to the initial binding of LPS to MD-2 , and the subsequent formation of the TLR4-MD-2-LPS complex [17] . Furthermore , the binding of all agonists to MD-2 induces a conformational change in MD-2 , but the fact that such structural change does not occur when an antagonist binds to MD-2 has provided the basis for the structure-function based development of new molecules to modify this protein's biological activity . In this context , the key observation has been that underacylated lipid A functions as an antagonist by occupying MD-2 and not inducing the conformational change in this protein required to trigger TLR4 activation and oligomerisation [29] . The most advanced antagonist in development is eritoran ( or e5564 ) . It is a synthetic di-glucosamine and lipid based molecule that is derived from the lipid A structure of the nonpathogenic LPS of Rhodobacter sphaeroides [30] . It binds and blocks the entry of LPS into MD-2's pocket and thereby prevents formation of the TLR4-MD-2-LPS cell surface receptor complex . Unlike lipid A , eritoran's antagonist activity depends solely upon electrostatic interactions with Lys122 and Lys125 on the β-7 strand of MD-2 . These published observations about the lipid A based bioactive site of LPS make our results surprising because they suggested that the presence of highly hydrophobic lipidic chains was a prerequisite for any agonist or antagonist to bind effectively to MD-2 [31] . However , our results show , for the first time , that the presence of lipidic chains is not an absolute requirement for an MD-2 antagonist or partial antagonist . Our molecular docking studies also show that these partially glycosylated dendrimers bound to human MD-2 and interfered with the electrostatic binding of:- ( i ) the 4′phosphate on the di-glucosamine of LPS to Ser118 on MD-2; ( ii ) LPS to Lys91 on MD-2; ( iii ) the subsequent binding of TLR4 to Tyr102 on the MD-2-LPS complex . These three residues line the hydrophilic entrance of MD-2's hydrophobic pocket ( Figure 5 ) . Additional polyvalent interactions between several other human MD-2 surface residues [i . e . , Arg96 , Ser98 , Arg106 , Lys109 , Thr112 , Asn114 and Thr116 ( Figures 5b & c ) ] and several of the partially glycosylated dendrimer's peripheral carboxylic acid branches ( Figure 9 ) enables it to position itself so that it always occludes the entrance to MD-2's pocket ( Figure 10 ) . This blocks the entry of the lipid chains of LPS into human MD-2's pocket and explains the molecule's antagonist activity . Our results also suggest that the interaction of the G3 . 5 partially glycosylated dendrimer with human MD-2 is specific . This conclusion is supported by the experimental biological data that only the TLR4 mediated pathway is affected by this molecule [6] . It should be noted that the dendrimer's surface glucosamines interact in a dynamic manner with several of the surface residues of MD-2 . This promiscuity of MD-2 is an important evolutionary feature and reflects the need of MD-2 to be able to recognise all of the diverse LPS structures synthesised by individual species of Gram negative bacteria . MD-2's promiscuity extends to the binding of endogenous host derived ligands that act as alert signals for tissue injury , such as enzyme derived fragments of hyaluronan from the extracellular matrix [32] . Our previous modelling studies suggested that a dendritic architecture was important for this molecule's biological activity [7] , [8] . In addition , our previous biological studies [6] were preceded by unpublished observations with smaller generations of partially glycosylated PAMAM dendrimers; these molecules did not have the desired biological activity . As the synthesis of higher generation dendrimers ( e . g . , G4 . 5 and higher ) is both expensive and leads to considerable polydispersity of the molecule , their biological activity was not tested . However , our findings , based upon the detailed modelling of the G3 . 5 partially glycosylated dendrimer , suggest that larger generations of these molecules should be biologically active because they explore a similar conformational space to the G3 . 5 partially glycosylated dendrimer . We have also been investigating the use of other polyanionic dendrimers with both different cores and branching units , but with the same carboxylated surface . Having successfully glycosylated different generations of these dendrimers , we have found that only some of them exhibit the biological activity required ( paper in preparation ) . We are coming to the conclusion that the main difference between the biologically active partially glycosylated dendrimers and biologically inactive molecules lies in their flexibility which , in turn , leads to changes in both their peripheral conformation and their surface electrostatic properties . Several years ago , we came to recognize the potential therapeutic importance for making polyvalent medicines because of our studies into the receptor-ligand interactions that mediate HIV-1 entry into cells [33] , [34] . We , and others , then realized that some aspects of cell surface mediated immuno-regulation depended upon co-operative electrostatic interactions between carbohydrates and proteins [35] , [36] . This has now been shown to be the case for the receptor-ligand interaction between TLR4 and LPS by the demonstration of an exponential increase in binding affinity [37] . Another example is the observation that four oligosaccharides are required for a functional immunological response to inhibit the LPS - DC SIGN cell surface interaction; N-acetyl-glucosamine-galactose-glucosamine and fucose-glucosamine-N-acetylglucosamine are required rather than N-acetyl-glucosamine or galactose on their own [38] . Our proof-of-concept studies have therefore demonstrated that it is possible to design a hyperbranched macromolecule with a pre-defined biological activity . The crucial advantage of our approach is that it avoids the expensive and complex chemistry associated with preparing oligosaccharides . This is because we can systematically modify the chemical functionality of the peripheral groups of dendrimers for a pre-defined biological activity using the simplified chemistry of monosaccharides to present aminosaccharides to cell surface receptors . As a result , medicinal grade product synthesis becomes a realistic goal for undertaking cost-effective clinical trials . Protein based medicines have already achieved this for macromolecules of >20 kDa by interacting with multiple cell surface receptors . Therefore , the use of dendrimeric structures of 2–5 kDa resolves two major obstacles that have impeded the therapeutic progression of synthetic macromolecules:- ( i ) excessive structural heterogeneity and poor control of molecular weight distribution; ( ii ) side-effects from in vivo activation of both complement and coagulation pathways [33] , [34] . Our approach also enables , for the first time , the adoption of existing molecular modelling software ( originally designed for studying protein-protein interactions ) for the study of the interactions of synthetic macromolecules with biologically important proteins . Taken together with our previous modelling studies , which rationalised the effects of pegylation on the structure and biological activity of both proteins and peptides [39]–[43] , we now believe that the principles and computational approaches that we have described could be applied more broadly to achieve proof-of-concept for other macromolecular structures . We have also provided new insights into the interactions between a synthetic hydrophilic macromolecule and a protein usually targeted by hydrophobic molecules . These observations , coupled with an increasing number of recently published studies that underscore the important functional role of electrostatic interactions in TLR-ligand interactions , should facilitate the design of novel macromolecules that are both chemically well defined and biologically useful for the therapeutic manipulation of important TLR based immuno-pathological pathways [44] . This is because the binding orientations amongst all TLRs are similar even though the residual interactions with their ligands are specific . There is also an increasing body of evidence to suggest that a network of hydrogen bonds controls the precise positioning of the ligand in a TLR , and that it is the molecule's precise positioning that determines the specificity of TLR-ligand complex mediated signalling events [45] , [46] . Established drugs are typically small molecules with a MWt of <500 Da . Biopharmaceuticals with a MWt of >20 kDa that are agonists of naturally occurring biological pathways are already on the horizon as important new medicines . Designing and delivering a novel generation of well defined chemically synthesised molecules with MWts of 2–5 kDa that mimic important biological properties is the new challenge for chemical-biologists . We propose that chemically optimised dendrimers with well defined chemical and biological properties ( which we also propose could be called “synthetic baby bios” ( SBBs ) ) can now be designed using the computational biology approaches described . These molecules could be useful for treating a spectrum of infectious , inflammatory and malignant diseases .
Hex and Patchdock are two different software packages developed for the study of protein-protein interactions . Their algorithms and scoring functions are different . Patchdock reads each structure and transforms it into Conolly surface representations [47] , [48] that include flat , concave and convex patches [49] . The complementarities between patches are assessed and then given a scoring function which takes into account both shape fitting and desolvation energy; i . e . , energy required to break molecule-solvent interactions . The results are cluster based on the RMSD values of each transformation [49] . The crystal structure of the mouse TLR4-mouse MD-2 complex ( PBD entry: 2z64 ) was used to evaluate whether the software could reproduce the structure of the complex . The mouse TLR4-mouse MD-2 complex was used because , at that time , there was no crystal structure for the human cell surface receptor complex . For the negative controls , the target used was the same as that used for the positive control docking experiments; i . e . , mouse TLR4 from the X-ray structure with PDB entry 2z64 . However , the ligand was not mouse MD-2 from the same crystal structure but instead human MD-2 from the two different crystal structures available ( i . e . , PDB entries: 3FXI and 2e56 ) . Docking for these negative control experiments was carried out using the same parameters as were used for the positive control experiments . In the first experiment to dock TLR4 with MD-2 , all of the oligosaccharides found in the original PDB entry file were removed . This resulted in a best fit that showed MD-2 interacting with TLR4 on the inside of the horseshoe ( Supplementary Figure S1-A; brown image ) rather than sitting on top of the TLR4 extracellular domain . When the change in docking target was changed to a TLR4 possessing all its oligosaccharides , this produced a good superimposition between the crystal structure and the docking result ( Supplementary Figure S1-B ) . A third experiment was performed keeping only the oligosaccharides inside the horseshoe . This also showed that the MD-2 from the crystal structure overlay the docked MD-2 ( Supplementary Figure S1-C ) . This result indicated that the presence of sugar molecules inside the horseshoe was crucial for the correct positioning of MD-2 in the TLR4-MD-2 complex . In the negative control experiments ( Supplementary Figure S2 ) , the lack of shape complementarity between the mouse and human regions critical for the TLR-4 and MD-2 interaction was used to confirm that a cross-species analysis ( i . e . , mouse TLR4 with human MD-2 ) did not reproduce the orientation of mouse TLR4 with mouse MD-2 . It was also notable that human MD-2 , taken from the crystal structure of the human TLR4-human MD-2 complex , gave a better solution that the use of human MD-2 complexed to lipid A from the crystal structure with PDB file entry 2z59 . These negative control experiments confirmed that Patchdock was suitable for more detailed studies once the target molecules had been correctly defined . The Hex software package was developed for protein and nucleic acid interaction studies . It also allows modelling of small ligand-protein interactions using rigid docking . Hex is a fast Fourier transform ( FFT ) docking correlations based programme that uses soft polar Fourier correlations to minimise the computational time required to explore the Cartesian space [50] , [51] . The protein's molecular surfaces are represented by an internal and an external “skin” that are each represented by a Fourier series , and comprise radial and spherical harmonic basis functions [50] . The electrostatics contribution is optional , when present , and is only taken into account in the final search [52]; it has a small weighting in the final scoring function . The resulting structures are ordered from lowest to highest energy , and then clustered with a 3 Å threshold for the main chain Cα - Cα RMSD values [52] . The same crystal structure of the mouse TLR4-mouse MD-2 complex ( PDB entry: 2z64 ) was used to evaluate whether the software could reproduce the structure of this complex . The influence of the “shape only” parameters versus “shape and electrostatics” parameters was also assessed . Since Patchdock revealed the importance of the presence of the oligosaccharides inside TLR4's horseshoe , these were considered in the docking process with Hex . The docking results of the mouse TLR4-mouse MD-2 complex showed that the Hex shape only protocol software was inadequate for describing the system correctly ( Supplementary Figures S3-A and S3-C ) . If a contribution from electrostatics ( even with a smaller weighting ) was taken into account , the docking reproduced the crystal structure albeit with a small deviation , but maintained the correct interaction site of TLR4 with MD-2 ( Supplementary Figure S3-B ) . These negative control experiments with Hex showed that the lack of shape complementarity between the human and mouse regions of MD-2 critical for TLR4-MD-2 complex formation ( i . e . , a cross-species analysis using mouse TLR4 and human MD-2 ) did not reproduce the orientation of mouse TLR4 in relation to mouse MD-2 ( Supplementary Figure S4 ) . Therefore , these modelling results were consistent with the biological observation that:- ( i ) pro-inflammatory cytokines are produced when HEK293 cells are transfected with human TLR4 and human MD2 and then stimulated with LPS; ( ii ) pro-inflammatory cytokine production does not occur when HEK293 cells are transfected with human TLR4 and mouse MD2 and then stimulated with LPS [53] . There is therefore a lack of shape complementarity between the human and mouse regions of MD-2 critical for the TLR4-MD-2 interaction . These observations led us to conclude that the Hex shape and electrostatics protocol was the most suitable for our further studies . For a better understanding of the contribution of electrostatics in the interaction of the dendrimer and the glycosylated dendrimer with the LPS recognition system , GRID software was used . GRID is a calculation based procedure [54] . It enables the determination of the energetically favourable binding sites on a molecule whose structure is known . GRID results can be visualized with the Gview application . It allows visualization of molecular interaction fields , GRID energy contributions due to atoms of the target , and molecular structures with distances , torsion , and dihedral angles . Glue is a GRID docking programme that is capable of finding potential interaction sites between a molecule , set as “target” , and a small molecule set as “ligand” . It requires the input of both the target and the 3D structures of the ligand . Its scoring function takes into account steric repulsion energy , electrostatics contribution , a dry parameter ( which accounts for hydrophobic energy ) , and an additional hydrogen bonding charge reinforcing parameter . The system used for Grid validation was human MD-2-lipid IVa ( PDB entry: 2z59 ) . The GRID results for the structure of the human MD-2-lipid IVa complex were very similar to the crystal structure of lipid A with human MD-2's hydrophobic pocket , with respect to both position and conformation ( Supplementary Figure S5 ) . In addition , the number of interactions between the residues of human MD-2 and the atoms of the partially glycosylated dendrimer ( as summarised from the 400 solutions of the docking study performed with Hex ) were plotted together with the number of interactions between the residues of human MD-2 and the single lipid A structure obtained from the crystal structure of its complex with MD-2 [PDB entry: 2z59] ( Supplementary Figure S6 ) . This highlighted:- ( a ) the hydrophobic nature of the binding of the acyl chains of lipid A to residues 27 to 37 of human MD-2; ( b ) the hydrophilic nature of the binding of the partially glycosylated dendrimer to the entrance of human MD-2's hydrophobic pocket {residues 84 to 127}; and ( c ) the competitive nature of the binding of these two molecules to the entrance of human MD-2's hydrophobic pocket . The negative control experiments used sucrose and maltose as ligands . In biological experiments , we first confirmed that these two ligands did not compete with the binding site of LPS on human MD-2 ( unpublished observations ) . They were then docked against the same target; i . e . , human MD-2 from the PDB file entry 2e59 . The results with maltose and sucrose confirmed that their binding sites were distant to the binding site of the lipid A of LPS to human MD-2 ( Supplementary Figure S7 ) . Based on these results , we considered GRID to be the most suitable software for our further studies . The structures of the proteins used for the docking studies were obtained through the RCBS website ( www . rcsb . org ) . The crystal structures of the mouse TLR4-mouse MD-2 complexed with eritoran ( PDB entry: 2z64 ) , and the human MD-2 complexed with lipid IVa ( PBD entry: 2e59 ) were used for validation of the docking software . The complex of human TLR4-human MD-2 ( PDB entry: 3FXI ) was used as a target for the docking studies; the non-glycosylated MD-2 without ligand was extracted from the crystal structure ( PDB entry: 3FXI ) and loaded into Maestro . Hydrogen atoms were then added and a molecular dynamics simulation performed with Desmond . The fully solvated system was built using the SPC solvation model and the size of the box determined automatically by creating a 10 Å buffer around the system being simulated . The molecular dynamics simulation was performed for 200 ps at 300K and 1 . 03 bar . These included structure minimisation and relaxation steps . The final structure was used as a target in the docking studies performed with Patchdock and Hex . Greater ( Molecular Discovery Ltd ) was used to generate the “ . kout” files for all of the molecules studied . Glue ( Molecular Discovery Ltd ) was used for docking studies . For the docking process itself , all available probes were selected to generate molecular interactions fields ( MIFs ) . A maximum of 100 binding sites were used with an energy cut-off of −100 kCal/mol . The maximum iterations value was set to 120 . For ligand flexibility , 5 rotatable bonds were allowed and the electrostatic term was included for the calculation of the interaction energies . Glucosamine was docked with the human TLR4-MD-2 complex ( PDB entry: 3FXI ) using several overlapping target volumes defined as a box with a side of 30 Å . Patchdock was used with all default parameters , except that the maximum surface overlap had to be changed to the lower value of −2 . The 20 lowest energy structures were saved as PDB files for further studies . The same protocol was used to dock mouse TLR4 with mouse MD-2 for validation , and to compare it with the crystal structure ( PDB entry: 2z64 ) . Once validated , the protocol was used to study the interactions of the G3 . 5 PAMAM dendrimer and the G3 . 5 partially glycosylated PAMAM dendrimer with:- ( i ) the human TLR4 - human MD-2 complex; ( ii ) human MD-2 . The docking studies carried out with Hex involved the use of two different protocols . The first was a shape based protocol . The second included both shape and electrostatics based protocols . All other parameters were the same for both protocols and were set to their default values . The scan steps were set to 0 . 75 with 2 substeps . The order of the docking correlation was set to 25 for the steric scan , and 25 for the fine search . The grid size was set to 1 and 100 solutions . Of these , the representative structures of the first 20 clusters were saved as PDB files for further analysis . The studies of the unmodified dendrimer and the partially glycosylated dendrimer against human MD-2 and the human TLR4-human MD-2 complex were performed with the shape and electrostatics protocol . Twenty different conformations of each biologically inactive and active molecule were used as ligands . For the visualisation of the outputs from the docking studies and to generate images , Discovery Studio Visualise v2 . 5 ( Accelerys ) was used . To process the 800 solutions resulting from the docking of the unmodified and modified molecules with human MD-2 , a rebol script implemented in VegaZZ was used . This script reported every atom in the dendrimer that was at a distance of <3 Å from an atom in human MD-2 . The resulting files were exported to Microsoft Excel and plotted as graphs . For energy scaling of the interactions , the number of interactions from each solution was multiplied by their energy value . The energy calculations were only performed for the Hex solutions . Initial conformations of the unmodified dendrimer and the partially glycosylated dendrimer were generated with X-PLOR using our “sequence to conformation” method [7] , [8] . These structures and the complex of human MD-2 with the partially glycosylated dendrimer were obtained from a docking solution using Hex imported into Maestro , and then set up for molecular dynamics simulation with Desmond . The fully solvated systems were built using the SPC solvation model , and the size of the box determined automatically by creating a 10 Å buffer around the system being simulated . The molecular dynamics simulations were performed for 4 . 8 ns at 300K and 1 . 03 bar . These included structure minimisation and relaxation steps . Snapshot structures were recorded every 4 . 8 ps . The contact surface areas between the partially glycosylated dendrimer and human MD-2 were determined from selected frames of the 4 . 8 ns trajectory using the Chimera software package ( www . cgl . ucsf . edu/chimera ) , having set the cut-off for van der Waal's interactions to 4 Å . The representative conformations of the unmodified dendrimer and the partially glycosylated dendrimer were obtained by trajectory analysis using Vega ZZ .
|
Dendrimers are well-defined branched symmetrical macromolecules . In biologically based experiments , we have shown that a generation 3 . 5 PAMAM dendrimer whose surface was modified with 8 surface glucosamine molecules inhibited TLR4 mediated cytokine inflammation in both primary human cells and a clinically validated rabbit model of tissue scaring . Molecular dynamics simulations also showed that these molecules had the flexibility , surface electrostatic charge , and hydrophilicity to make them therapeutically useful antagonists . Central to the TLR4-MD-2-LPS cell surface complex is binding of lipopolysaccharide ( LPS ) to soluble MD-2 . We now show that dendrimer glucosamine forms co-operative electrostatic interactions with residues lining the entrance to MD-2's hydrophobic pocket . These interactions block the entry of the lipid chains of LPS into MD-2's hydrophobic pocket and prevent complex formation . We have therefore defined the first nonlipid-based synthetic MD-2 antagonist using both animal model based studies and molecular modelling studies of a whole dendrimer with its target protein . Using this approach , it should now be possible to computationally design additional macromolecular dendrimer based antagonists for other Toll Like Receptors . They could be useful for treating a spectrum of infectious , inflammatory and malignant diseases .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[
"computer",
"science",
"chemistry",
"biology"
] |
2011
|
Partially Glycosylated Dendrimers Block MD-2 and Prevent TLR4-MD-2-LPS Complex Mediated Cytokine Responses
|
Viral gene expression varies significantly among genetically identical cells . The sources of these variations are not well understood and have been suggested to involve both deterministic host differences and stochastic viral host interactions . For herpesviruses , only a limited number of incoming viral genomes initiate expression and replication in each infected cell . To elucidate the effect of this limited number of productively infecting genomes on viral gene expression in single cells , we constructed a set of fluorescence-expressing genetically tagged herpes recombinants . The number of different barcodes originating from a single cell is a good representative of the number of incoming viral genomes replicating ( NOIVGR ) in that cell . We identified a positive correlation between the NOIVGR and viral gene expression , as measured by the fluorescent protein expressed from the viral genome . This correlation was identified in three distinct cell-types , although the average NOIVGR per cell differed among these cell-types . Among clonal single cells , high housekeeping gene expression levels are not supportive of high viral gene expression , suggesting specific host determinants effecting viral infection . We developed a model to predict NOIVGR from cellular parameters , which supports the notion that viral gene expression is tightly linked to the NOIVGR in single-cells . Our results support the hypothesis that the stochastic nature of viral infection and host cell determinants contribute together to the variability observed among infected cells .
Cell-to-cell variability is an important factor in cancer , development , evolution , host-pathogen interactions and other key biological processes [1–3] . The variability observed among single cells mainly arises from deterministic factors , i . e . preexisting molecular regulatory mechanisms [3 , 4] . In the context of viral infections , it was suggested that stochastic interactions between a virus and individual host cells could contribute to variability in the outcome of infection in the entire infected organism [5–9] . Much of the variation in the outcome of infection can be attributed to the specific cell state prior to infection [10] . Here , we provide evidence that the viral gene expression level also depends on the actual number of viral genomes initiating the infection process . Genetically barcoded viruses are used for studying cellular clonality ( see for example: [11–13] ) ; however , only a few studies have used genetic barcoding of viral genomes to study viral properties . Barcoded RNA viral genomes were used to identify bottlenecks in viral diversity , both inside the infected host [14 , 15] and during transmission among hosts [16] . Thus , barcoding of viral genomes can be a useful tool in studying bottlenecks during viral replication , even on the single cell level [17] . Herpes simplex virus 1 ( HSV-1 ) , a large DNA virus , is a very common human pathogen that causes significant morbidity throughout the world . HSV-1 is part of the large family of herpesviridae , and its replication has been studied as a model for the entire family of viruses . To replicate , naked herpes genomes enter the nucleus . Upon entry , the naked viral DNA associates with host histones to form nucleosomes . These chromatin structures are regulated by host histone modifying enzymes and are essential both for the lytic and the latent viral infection pathways [18 , 19] . Recent studies suggest that these interactions are cell type specific [20 , 21] . Interactions between the viral DNA , the tegument protein VP16 and host factors determine the probability of initiating immediate early gene expression [22] . Immediate early proteins activate expression of early and late genes and counteract host defense mechanisms . Both intrinsic and innate immunity are inhibited by the viral immediate early protein , ICP0 [23] . The requirement for ICP0 function varies between different cell types [24 , 25] . Thus , HSV-1 closely interact with the host cells , and specific mechanisms in the host cells can modify the outcome of the infection . Following early gene expression , viral replication initiates in specific domains known as pre-replication compartments [26] . These small structures grow in size , move and coalesce to form replication compartments ( RCs ) [26 , 27] . On the other hand , only one parental genome can be found in each RC [28 , 29] . The number of RCs was reported to be limited [26 , 30] . A bottleneck , limiting the number of incoming herpes genomes that are expressed and replicated , was observed; this number was estimated to be less than 10 per cell ( even in multiplicity of infection 100 ) [29 , 31 , 32] . To test if viral gene expression is dependent on the number of incoming viral genomes replicating ( NOIVGR ) per cell , we developed an HSV-1 genetic barcode system that estimates the NOIVGR at the level of single cells . Our system is based on DNA barcoding of mCherry expressing HSV-1 recombinants and identification of the progeny barcoded recombinants derived from a single cell . We determined that the mCherry fluorescence levels are surrogates for viral gene expression and found fluorescence levels to correlate with the NOIVGR . We developed a model that predicts the NOIVGR according to cellular parameters . We conclude that cellular determinants and stochastic effects of NOIVGR determine the outcome of infection in individual cells .
To study single cell variables that determine herpesvirus gene expression , we developed a system based on isogenic genetically-tagged recombinants of HSV-1 strain 17 . Fourteen unique sequences were inserted into a pOK11 plasmid [32] downstream to the mCherry gene ( see Methods ) . Using homologous recombination , these short sequences and the mCherry gene were inserted into HSV-1 genomes , resulting in 14 DNA-barcode tagged viruses . All 14 recombinants showed comparable growth curves ( S1 Fig ) and were verified by qPCR . The genome from each virus can be specifically identified by the barcode sequence ( S2 Fig ) . We hypothesized that the measured mCherry fluorescence expressed from the CMV promoter cloned into the viral genome is a surrogate for the expression of native viral genes . To test this hypothesis , we sorted Vero cells infected with barcoded viruses at MOI 10 or 100 into three populations according to their fluorescence ( 30% lowest , 40% middle and 30% highest of the total population ) . Viral gene expression levels of immediate-early ( ICP4 , UL54-ICP27 ) , early ( UL5 , UL29 ) and late ( UL19 , US7 ) genes were measured by RT-qPCR from each population . We compared expression of the lowest and highest populations to the middle subpopulation ( Fig 1A and 1B ) . As expected , in the highest fluorescence population all viral genes were expressed at a higher level than in the middle subpopulation . Similarly , in the lowest fluorescence population , viral genes were expressed at lower levels . To ensure that the correlation between the mCherry expression from the CMV promoter to the viral gene expression is maintained throughout the infection process , Vero cells were infected with a dual color expressing virus ( OK21 ) which carry the mCherry expressed from the CMV promoter and the mVenus ( a yellow fluorescent protein ) fused into the late gene UL25 under its native promoter . We monitored the infection , and at representative time points , we estimated the mCherry and YFP fluorescence levels ( Fig 1C and 1D ) . Our results indicate a significant continuous correlation between the levels of the mCherry and YFP expressed from the late viral promoter . Together with RT-qPCR results , we conclude that the fluorescence expressed from the CMV promoter in viral genomes is a surrogate for viral gene expression during the entire infection process . To test if the relative mCherry fluorescence level measured at a given point in time represents the relative fluorescence level throughout the infection , we followed infected Vero cells for 16 hours under a fluorescence microscope . The movies obtained ( Fig 2A , S3A Fig and S1 and S2 movies ) suggest that cells expressing fluorescence early , maintain a high level of fluorescence throughout the infection . To quantify this phenomenon , we plotted the fluorescence level from each cell at each time point ( Fig 2B and 2C and S3B and S3C Fig ) . We categorized the cells according to the fluorescence level at 4 hours post infection ( HPI ) ( 30% of the cells with the highest or lowest fluorescence ) . In most of the cells , the level of fluorescence did not change throughout the infectious cycle ( Fig 2D and 2E and S3D and S3E Fig ) . At MOI 100 , less than 10% of the cells that were in the lowest fluorescence group at 4 HPI , switched to the highest fluorescence group at 16 HPI , and vice versa . At an MOI 10 , this distinction was less prominent , due to the low fluorescence level at 4 HPI . We noticed that at 16 HPI some of the cells showed cytopathic effects , and moved from the focal plane , a phenomenon that may account for some of the movements between the groups . These results indicate that the relative fluorescence level throughout the infection is a consistent parameter of an infected individual cell . The implication is that relative fluorescence levels can be represented by a single measurement during the infection , as in our experimental system described below . To ensure that we identify all viral genomes expressing and replicating in a given cell , we monitored color variation of cells during infection with three different fluorescence expressing recombinants . The recombinant viruses ( OK11 , OK12 and OK22 ) express mCherry , EYFP and mTurq2 , respectively under the CMV IE promoter [32] , [33] . Vero cells were infected with an even mixture of the three recombinants at MOI 100 and were visualized for 16 hours under a fluorescence microscope . Approximately 5 HPI ( the time it takes for all fluorescent proteins to express , fold and accumulate to detectable levels ) , each cell obtained a specific hue that remained relatively stable the entire infection time ( Fig 2F and 2G , S3 Movie ) . Notably , while the cell became brighter in time , the proportion of the different colors remained similar . These results suggest that expression is initiated and maintained throughout the infection from the same input viral genomes . As we previously showed that expressing genomes are also the ones replicating [31] , we conclude that most incoming viral genomes , that initiate replication , are detectable . We hypothesized that cell-to-cell variation might result from differences in the number of incoming viral genomes replicating ( NOIVGR ) in each cell . To test this possibility , we developed a method to assess the NOIVRG in individual cells . Cells were infected with a mixture of fourteen barcoded viruses ( Fig 3A ) . The mixture contained equal quantities of plaque forming units ( PFU ) from each virus . The cells were infected at MOI either 10 or 100 ( Fig 3B ) . At 3 HPI , the cells were collected and sorted according to their fluorescence intensity ( Fig 3C ) . Individual cells were sorted onto a pre-seeded 96-well plate ( Fig 3D ) and were allowed to develop into infectious centers . At 144 HPI , the infectious center spread to most of the cells in the well ( Fig 3E ) . The contents of each well were collected , lysed and analyzed with qPCR , to determine the identity of the progeny barcodes ( Fig 3F ) . First , we infected Vero cells and isolated single cells as described above . The number of barcodes detected in each well at MOI of 100 ranged between 1 and 13; and at MOI of 10 , between 1 and 7 ( Fig 4B and 4C ) . These results suggest that the barcodes detect the NOIVGR in the initial infected cell; and corroborate our previous estimate of the number of expressed herpes genomes inside a cell [31 , 32] . We hypothesized that progeny barcode viruses derived from the infectious center represent the replicated parental viruses in the original single Vero cell . We ensured that barcoded recombinants have similar growth properties ( S1 Fig ) . Fig 4A shows that at each MOI all fourteen barcodes can be detected by qPCR ( a collective output from all the wells tested ) . Moreover , the distribution of all the barcodes after an MOI 100 was similar . At MOI 10 , the distribution was noisier , probably reflecting the low number of barcodes per cell at this MOI , and the subsequent stochastic effects . Taken together , these results suggest that the recombinant viruses outcompete each other randomly per cell . Sorting the cells enables measuring the forward scatter ( FSC is usually considered as a measure of cell size ) , side scatter ( SSC is usually considered as a measure of cell granularity ) and fluorescence of each individual cell . We evaluated the relationship of these parameters to the NOIVGR ( S5 Fig ) . Our initial results indicated that MOI and fluorescence were dominant predicting factors for the number of parental genomes that replicated in Vero cells . To characterize the role of MOI in predicting the NOIVGR , we performed all experiments with MOI of 10 or 100 . We found that the average number of replicating viral genomes , as indicated by the mean number of barcodes detected by qPCR , is dependent on MOI ( for MOI of 10 , the mean was 3 . 4 , and for MOI of 100 , 7 . 6 , p< 0 . 001 ) ( Fig 4B and 4C ) . To evaluate the correlation between fluorescence and the NOIVGR , we sorted cells according to their fluorescence: low , intermediate and high ( S4A Fig ) . Cells expressing lower fluorescence replicated a smaller number of viruses than did those expressing intermediate or higher fluorescence ( Fig 4B and 4C and S1 Table 1 ) . At MOI 10 ( Fig 4B ) , differences between the high fluorescent cells and cells of either low or intermediate fluorescence were statistically significant ( p < 0 . 001 ) . Moreover , the difference between the low and intermediate groups was also statistically significant ( p < 0 . 04 ) . At MOI 100 ( Fig 4C ) , differences between low fluorescence and the other two categories were statistically significant ( p < 0 . 001 ) ; however , the difference between the intermediate and high groups was minimal . We accumulated 64 cells infected at MOI 10 , and 73 cells at MOI 100; and plotted the number of barcoded progeny against the level of fluorescence ( Fig 4D and 4E ) . We calculated the Pearson correlation between these parameters for each MOI ( 0 . 57 and 0 . 50 for MOI 10 and 100 , respectively , p < 0 . 001 for both MOIs ) . Taken together , these results indicate that the MOI and the level of fluorescence emitted by infected cells enable estimating the NOIVGR in single cells with a high degree of confidence . To test whether the correlation between fluorescence and the NOIVGR can also be detected in human cells , we infected immortalized human foreskin fibroblasts ( HFF ) with the mixture of the 14 barcoded recombinants . As was observed for the Vero cells , all the barcodes were detected in comparable amounts from the sorted individual cells ( Fig 5A ) . The average number of barcodes detected per cell in HFF was slightly different than observed in Vero cells ( lower at MOI 100 and higher at MOI 10; S1 Table ) . The infected cells were sorted according to their fluorescence: low , intermediate and high ( S4B Fig ) . HFF expressing lower fluorescence replicated a smaller number of viruses than did those expressing intermediate or higher fluorescence ( Fig 5B and 5C and S1 Table 1 ) . At both MOIs , differences between low fluorescent cells and cells of either intermediate or high fluorescence were statistically significant ( for MOI 10 P < 0 . 005; for MOI 100 P < 0 . 001 , Fig 5B and 5C , respectively ) . The difference between the intermediate and high groups was minimal . The number of barcodes per cell from 66 cells infected at MOI 10 and 69 cells infected at MOI 100 were plotted against the level of fluorescence ( Fig 5D and 5E ) . We calculated the Pearson correlation between these parameters for each MOI ( 0 . 55 and 0 . 47 for MOI 10 and 100 , respectively , P < 0 . 001 for both MOIs ) . Our results show that the correlation between fluorescence level and the NOIVGR can be detected in HFF . We conclude that this correlation is probably not cell type specific and is a more general characteristic of HSV-1 infection . We speculated that cell to cell variance in viral gene expression ( i . e . fluorescence from the viral genomes ) depends on the basal gene expression of the specific cell . To test this idea , we compared the level of viral fluorescence to expression from a cellular housekeeping gene ( HKG ) promoter . HeLa cells harboring the green fluorescent protein ( GFP ) under phosphoglycerate kinase ( PGK ) promoter were infected with one of the barcoded viruses . Surprisingly , not only was there no positive correlation between the intensity of the host fluorescence ( GFP ) with the viral fluorescence ( mCherry ) , ( as was shown in Fig 1C and 1D for two viral expressed fluorescent proteins ) but in many cases the very highly expressing GFP cells had very little viral expression ( Fig 6A ) . First , we verified that the GFP levels were not affected in the presence of viral infection . Fig 6B shows the average mCherry fluorescence and GFP fluorescence of uninfected green HeLa cells ( gHeLa ) compared to infected cells at MOI 100 or 10 . The results indicate that there was no significant change in the GFP levels during infection . Therefore , the relative levels of GFP expression per cell at the end of experiments represent the relative basal levels of GFP expression prior the infection . We analyzed images at 10 HPI of gHeLa cells infected with one of the barcoded viruses ( three different recombinants were used ) at MOI 10 or 100 . We determine the relative fluorescence ( both for mCherry and GFP ) for each cell from all eight images and plotted the frequency of the relative fluorescence ( combined data from all frames analyzed ) as a Heatmap ( Fig 6C and 6D for MOI 10 and 100 , respectively ) . Our results indicate that the distribution of the two colors among cells is not random ( Chi square test , P<0 . 005 ) . The proportion of cells with low mCherry levels ( left column ) with high GFP levels ( upper rows ) is significantly higher from the expected distribution , in both MOIs ( Fig 6C and 6D ) . Thus , cells with initial high GFP levels are less likely to express high levels of viral gene expression . These results suggest an inverse association between viral gene expression and the basal levels of host HKG expression . To test whether this association is specific to the promoter tested , we used a set of 12 randomly chosen cell clones from the LARC ( library of annotated reporter cell-clones ) . These cells express the yellow fluorescent protein ( YFP ) gene fused to different host genes ( the specific genes are listed in the methods ) and is expressed from the native promoter in its chromosomal position [34] . The 12 LARC were infected with one of the barcoded viruses ( each cell line separately ) . In all cells , the YFP signal was not eliminated during infection F[35] . We analyzed images at 8 HPI following infection at MOI 10 or 100 . We accumulated the relative levels of YFP and mCherry from all the cell-lines tested and show the results as Heatmaps for MOI 10 and 100 ( Fig 6E and 6F; respectively ) . Our results indicate that the distribution of the two colors among cells is not random ( Chi square test P<0 . 05 ) . The obtained data show a similar phenomenon as was observed in the gHeLa cells , although it was less distinct , probably because of variability among the LARCs . These results support the inverse association between viral gene expression and the basal levels of host gene expression . To further study the NOIVGR in different cell types , we repeated the single cell analysis ( as described above ) on gHeLa cells . We accumulated 50 cells infected at MOI 10 , and 67 cells at MOI 100 . All fourteen barcodes were detected at each MOI ( Fig 7A ) . The barcode distribution in the gHeLa cells was noisier than in Vero or HFF cells ( compare with Figs 4A and 5A ) . The noisier distribution of barcodes suggests a lower number of barcodes per gHeLa cell . Indeed , we found a lower mean of barcodes per gHeLa cell in both MOI 10 and 100 ( 2 . 1 and 4 . 4 , respectively , Fig 7B and 7C and S1 Table ) compared with Vero and HFF cells . Single cell HGK expression levels ( GFP fluorescence ) were not found to correlate significantly with the NOIVGR ( Fig 7D and 7E ) or with viral gene expression ( S6 Fig ) . We assume that this is due in part to the small sample size and to the small diversification in the GFP levels . We sorted HeLa cells according to their mCherry fluorescence: low , intermediate and high ( S4C Fig ) and found that the dependence of the NOIVGR on the viral gene expression was maintained in the gHeLa cells . The correlation was clear in MOI 100 ( p<0 . 001 , Fig 7G ) . At MOI 10 , the low number of barcodes per cell yielded a result with less statistical significance ( p<0 . 01 , Fig 7F ) . We conclude that while different cell types have different mean NOIVGR per cell; the correlation of this number to viral gene expression levels is likely to be cell type independent , and thus represents a fundamental process during viral infection . To assess the predictive value of single cell parameters in estimating the NOIVGR per individual cell , we developed a linear model based on the experiments described above . We extracted a linear relation between the number of barcoded progeny ( θ ) to MOI and single cell parameters , as measured by the cell-sorter . We used 137 single Vero cells from three independent experiments , 135 single HFF from three independent experiments and 117 single HeLa cells from four independent experiments to calculate the linear equation , using the SPSS linear regression tool . We defined θ as the dependent variable and included MOI , FSC ( cell size ) , SSC ( cellular granularity ) and mCherry fluorescence as variables for analysis ( Eq ( 1 ) ) . The model suggests that the coefficients of MOI , fluorescence , FSC and SSC are significant ( P < 0 . 001 , for MOI , fluorescence and FSC; and P < 0 . 05 for SSC ) . To test the linear model , we calculated the Pearson correlation between the predicted values obtained from the model and the measured values for each cell . First , all 389 single cells used to develop the model were compared to their predicted values from the model ( Fig 8A and S6A , S6E and S6I Fig ) . The calculated Pearson correlation ( 0 . 70 , P < 0 . 001 ) suggests that while the model predicts the NOIVGR , other single cell parameters should be included to obtain a better estimate . To test the predictive power of this system , we constructed a linear regression model from either of the cell types ( each alone ) and used these models to asses all cells ( Fig 8B–8D and S6B–S6D , S6F–S6H and S6J–S6L Fig ) . The calculated Pearson correlation from all models on all cell types separately and together are presented in Fig 8E . The calculated correlations obtained from each of the models are comparable ( All Pearson correlations were above 0 . 60 , except for the HHF based model predicting all cells which was 0 . 59 ) . The values of the Pearson correlations indicate that our model can identify the cells that have the highest ( or lowest ) NOIVGR in any experiment . However , Pearson correlation is not a good indicator for the ability of the model to identify the actual NOIVGR per cell . To test the capability of the model to retrieve the actual NOIVGR , we calculated slopes of the best-fitted trend-lines ( forced to intercept at 0 ) from these models ( Fig 8F ) . We found that the model obtained by either Vero cell experiments or HFF experiments over-estimated the NOIVGR in HeLa cells ( slopes ~1 . 5 ) . Similarly , the model obtained by HeLa cell experiments under-estimated the NOIVGR in Vero cells , HFF and all cells ( slopes ~0 . 5 ) . These differences are probably the result of the differences in the average NOIVGR observed per cell type . Thus , our model can predict NOIVGR in a single cell , according to MOI and single cell parameters determined by the cell sorter . Further , observations of correlations among parameters of different cell types suggest that basic parameters of our models are fundamental to the infection process , and have a predictive property for different cell types .
Cell-to-cell variations during viral infection contribute to significant differences in the outcomes of infection [36–39] . To study sources of single cell diversity during herpesvirus infection , we constructed fourteen barcoded , fluorescence-expressing viral recombinants . The expressed fluorescence from infected cells was found to be a surrogate for viral gene expression . We compared fluorescence levels in individual cells to the number of incoming viral genomes replicating ( NOIVGR ) in these cells . In the three cell types we examined , the NOIVGR correlated with fluorescence expressed from the viral genome , even though differences in the mean NOIVGR among the cell types were observed . Unexpectedly , our results suggest that cells with high basal gene expression levels are less permissive for viral infection . We also developed multiple linear regression models that predict the NOIVGR in a single cell according to cellular parameters , suggesting that the correlations we observed can be extrapolated to other experimental systems . We followed the fluorescence level of mCherry expressed from the CMV promoter cloned into the viral genome . We found that fluorescence expressed by cells coincides with viral gene expression throughout the entire infection process ( Fig 1 ) . Further , we showed that infected cells maintained their initial relative fluorescence ( Fig 2 ) . Taken together , we used the fluorescence level of infected cells at a single time point as an indicator of viral gene expression during infection . We detected small subpopulations of cells in which relative fluorescence changed during infection . Most cells that changed from the highest expression group to the lowest , probably represent cytopathic effect in which the cells lose their focal point and fluorescence ( evidence for these cells can be seen in Fig 2B ) . Cells that changed from the lowest expression group to the highest , are cells in which the fluorescence initiated relatively late , but once initiated , it accumulated fast . This is unlikely to be due to more viral genomes that establish infection and replication at later time points , as no such cells were identified with the three color infections ( Fig 2F and 2G and S3 Movie ) . We therefore speculate that in these cells , a threshold level of viral proteins may be needed to achieve efficient expression and replication , as was previously suggested for reactivation process [40] . To identify factors that influence viral gene expression in an infected cell , we obtained the NOIVGR and cellular parameters from the cell sorter . Fluorescence , as measured by the sorter , was used as a surrogate for viral gene expression . We consistently observed a correlation between fluorescence and the number of barcoded progeny viruses . The number of DNA copies of a gene affects the level of gene expression [41 , 42 , 43] . In this case , the higher the number of viral genomes initiating expression , the more gene expression ( fluorescence ) is observed . On the other hand , NOIVGR in a given cell does not necessarily represent the actual viral genome copy number during lytic replication , as genome number probably changes rapidly in an asynchronous manner among cells [26 , 28 , 44] . The correlation between NOIVGR and gene expression levels ( fluorescence ) may reflect the initial expression level from the genomes . If viral immediate-early genes are expressed efficiently , they are likely to create conditions that favor expression and replication of co-infecting viral genomes [45–48] . This immediate-early gene hypothesis suggests that viral gene expression may depend on pre-existing single cell determinants . On the other hand , the copy number hypothesis represents a model in which the NOIVGR is stochastic and determines the outcome of infection . These hypotheses are not mutually exclusive and we suggest that both contribute to the outcome of infection in a single cell . In all the cell types tested , the distribution of the barcode number per cell ( Figs 4 , 5 , and 7 ) approximates a Poisson distribution for almost any sub-population tested ( except for the intermediate HeLa cells at MOI 10 ) . Approximation to the Poisson distribution suggests that each cell contains a different number of viral genomes independent of other cells in the population . With the 14 barcodes , we can detect up to 14 different parental genomes . However , we cannot exclude the possibility that several parental genomes with the same barcode may enter a single cell . To estimate the difference between the numbers of output barcodes to the number of genomes initiating replication , we calculated the expected values of NOIVGR given the observed number of barcodes ( S2 Table ) . As expected , the overall average number of barcodes detected per cell is lower than the actual NOIVGR . For example , we calculated that for Vero cells at MOI 10 , cells with four barcodes have an expected value of 4 . 5 NOIVGR , meaning that for Vero cells with four barcodes infected at MOI 10 , the average number of genomes replicated in these cells is 4 . 5; as none of the cells can have less than four genomes , at least half of the cells will be replicating four genomes . Our calculations suggest that the barcode system provide a close estimate to the actual number of genomes initiating replication , especially for the cells in which a low number of barcodes was detected per cell . Similarly , at specific conditions , in which the average number of barcodes per cell is lower than other condition ( for example: MOI 10 and 100 , respectively ) , the results obtained by the barcodes are closer to the true NOIVGR . Previously , we found that on average , only a limited number of incoming herpes genomes are expressed in each cell . These results were obtained using three recombinant viruses , each expressing a different fluorescent protein ( As in Fig 2F and 2G ) . Cells infected with an equal mixture of the three recombinant viruses were imaged and analyzed for the presence of fluorescence at each color . A mathematical model predicted that on average less than ten genomes are expressed per cell [29 , 31 , 32] . In the current work , we measured the number of barcodes in progeny viruses originating from single cells , and used these data to estimate the NOIVGR in the original isolated single cell . It is important to note that in the previous work , we measured only the average numbers of genomes and not the distribution , thus we could not estimate the number of genomes expressed per cell . With the barcode system we observed up to 13 barcodes in one cell , indicating even a higher number of genomes initiating replication ( S2 Table ) . Despite differences between the assays , the results are consistent , indicating that the average number of replicating parental genomes is similar to the average number of expressed parental genomes . Everett and colleagues , and our work , suggested that each RC originates from a single parental genome [28 , 29] . However , Phelan et al . suggested that the number of RCs is higher than the number of expression foci late after infection [30] . Our data suggest that the RCs observed late during replication by Phelan et al . originated from genomes already replicating within the cell . The combination of single cell analysis with the barcode genetic system provides a new method to study the effect of bottlenecks imposed by single cells on the genetic diversity of DNA viruses . The limited capacity of single cells to replicate multiple RNA viral genomes was recently described as a bottleneck that limits viral diversity . Interestingly , the investigators of that study found that the reduced genetic diversity arising from a single cell correlates with the viral yield from that cell [49] . We suggest that barcode system serves as a measurement of the bottleneck of viral diversity . Similar to the situation with RNA viruses , our results suggest a positive correlation between gene expression of DNA viruses and genetic diversity ( barcodes ) . Thus , single cells are an inherent limiting factor for viral diversity . To test the effect of cell basal expression levels on viral gene expression , we measured fluorescence obtained from the housekeeping gene PGK promoter . The PGK promoter is commonly used to provide stable expression in mammalian cells [50] . Unexpectedly , basal expression levels did not positively correlate with viral gene expression . Further , cells with high host gene expression were less likely to express high viral expression ( Fig 6A–6D ) . To investigate whether this phenomenon is unique to the PGK promoter , we tested a dozen more host promoters ( these promoters are considered active in the cell at all times ) . The analysis of these twelve promoters together showed a similar phenomenon ( Fig 6E and 6F ) in which the high host expression levels are less likely to be infected efficiently . Similarly , but less predominant , we observed that cells with high viral gene expression are likely to have low host gene expression . In contrast , two promoters expressed from the same virus are mostly positively correlated ( Fig 1C and 1D ) . In addition , GFP expression from the HSV-1 genome is not inhibitory for viral acute replication [51] , suggesting the effect we observe is not due to inhibitory effect of GFP ( or its YFP derivative ) on viral replication . Taken together , we suggest that cells with higher basal level of host gene expression , also have higher levels of restriction factors ( i . e . intrinsic immunity proteins reviewed in [23] ) , which , by definition , are genes that can be found in the cell at all times . The consequence is a more effective inhibition of viral gene expression . Indeed , a work from our lab suggests that a small increase in known host restriction factors against HSV-1 can result in a decrease in the average number of herpes viral genomes that establish infection per cell [52] . Here we observe differences in the average NOIVGR between the different cell types we tested . While the two immortalized cell types ( Vero and HFF ) show a similar average number , the cancerous HeLa cells show a significant decrease in the average number of genomes . As HeLa cells genome carries many abnormalities and gene duplications [53] , it is possible that some host restriction factors are overexpressed in these cells , which may explain the reduced number of viral genomes expressed . On the other hand , due to genetic abnormalities , HeLa cells may have deleted , mutated or downregulated genes that are required for efficient initiation of viral replication . All viruses were grown and tittered on Vero cells before they were used in the various experiments . Indeed , this potentially could be a source for differences between cell types , as some of the tegument proteins in the virion originate from the host cell [54] . However , as the average NOIVGR of HFF is similar to the average of Vero , and the efficiency of plating is also similar between all three cell-types , we assume that this experimental parameter has only minimal effect ( if any ) on the results presented . We developed a model based on all single cell experiments that estimate the number of parental viral genomes that replicate in any given cell . This model was constructed as a multiple linear regression with all possible explanatory variables ( MOI , fluorescence , FSC and SSC ) . Eq 1 indicates that all measured parameters are positively correlated to the NOIVRG . The effects of MOI and viral fluorescence on the NOIVGR were discussed above . FSC ( cell size ) and SSC ( cell granularity ) can distinguish between the different cell types tested . For example , HFF and Vero are larger cells than HeLa ( higher FSC , S5 Fig ) . Taken together , all the parameters allow the prediction of NOIVGR depending on cell type , fluorescence and MOI . The model predicted the results of all cells together , as well as each cell-type separately ( Fig 8 and S7 Fig ) . We developed similar models based on results obtained from only one of the three cell-types . Remarkably , these models had similar Pearson correlation values for each experiment , as did the original model , although a bias was observed among the cell-types , which corresponded to the difference in mean NOIVGR measured . We conclude that the correlation between the NOIVGR and the cellular parameters measured is a robust phenomenon and may be extrapolated to different cell-types . In conclusion , we developed an experimental system that identifies single cell variation during HSV-1 infection . Further , this system provides evidence to the connection between viral gene expression and viral diversity . We suggest that the high basal level of gene expression in a given cell is a key determinant for reducing the probability for initiation of infection from herpes genomes . Our results support the hypothesis that host cell determinants , in combination with the stochastic nature of viral infection , contribute to the outcome of infection .
The single cell experiments were performed with either green monkey kidney cells ( Vero cells , ATCC CCL-81 ) , immortalized human foreskin fibroblasts ( HFF ) or human cervical cancer cells ( HeLa cells ) . These HFF cells were immortalized by hTERT transfection . The immortalized HFF cells were a kind gift from the Sara Selig . HeLa cells used in this work were infected with a lentivirus carrying the GFP gene under the mouse PGK promoter , derived from the pMSCVpuro plasmid ( Clontech ) . These HeLa cells were a kind gift from the E . Bacharach lab . The HeLa cells were sorted into single cells to ensure all experiments were done from a single clone , and that GFP expression differences are not due to insertion site . Cells were grown with Dulbecco's Modified Eagle Medium ( DMEM X1; Gibco ) , supplemented with 10% Fetal Calf Serum ( FBS; Gibco ) and 1% penicillin ( 10 , 000 units/ml ) and Streptomycin ( 10mg/ml ) ( Biological Industries Israel ) at 37°C and 5% CO2 . The HeLa cells were grown in the presence of G418 ( 0 . 5mg/ml Gibco ) to maintain fluorescence expression . 12 randomly chosen clones from the LARC ( library of annotated reporter cell-clones ) were a kind gift from Uri Alon . These cells originated from the non-small cell lung carcinoma cell-line , H1299 . Each of the clones has the yellow fluorescent protein ( YFP ) gene fused to different host gene and is expressed from the native promoter in its chromosomal position [34] . The 12 clones used in this study are: NASP , DNMT1 , SUMO1 , CIRBP , LMNA , LMNB1 , ACTN4 , H2AFV , FBL , RPS11 , MSN and TOP1 . Cells were grown in RPMI supplemented with 10% FBS ( Gibco ) and 1% penicillin ( 10 , 000 units/ml ) and Streptomycin ( 10mg/ml ) ( Biological Industries Israel ) . at 37°C and 8% CO2 . All viruses are derivatives of HSV-1 strain17+ . Viral recombinants OK11 , OK12 and OK22 carry a single fluorescent protein ( mCherry , EYFP and mTurq2 , respectively ) with a nuclear localization tag under the CMV promoter between UL37 and UL38 genes as described previously [32 , 33] . Viral recombinant OK21 was created by a cross between OK11 and OK19 ( a HSV-1 strain17+ containing the mVenus gene fused in frame within the UL25 gene after the 50th amino acid of the viral protein similar to what was previously described [55] ) . Fourteen 16-nucleotide sequences were randomly generated , and the sequence of BclI restriction site ( GATC ) was added to each . The resulting fourteen 20bp sequences were synthesized by Integrated DNA Technologies , Inc . ( IDT ) and cloned into pOK11 plasmids ( an mCherry with a triplicate nuclear localization sequence ( 3xNLS ) under the CMV Promoter in between homology regions of HSV-1 UL37 and UL38 genes ) as was previously described [32] . The 14 generated plasmids were cotransfected ( TRANSFECTENE@ PRO , Biontex ) into Vero cells with purified HSV-1 strain 17 DNA [56] . Each recombinant was plaque purified according to fluorescence and analyzed with qPCR to verify the insertion . The growth properties of each recombinant were compared to the parental strain , using a single-step growth curve analysis ( S1 Fig ) . All viral stocks were cultured and tittered on Vero cells . Vero , HFF or gHeLa cells were infected with an equal mixture of barcoded viruses at different MOIs , as specified in each experiment . At 3 HPI , the cells were trypsinized and kept on ice until sorting . The cells were sorted with Astrios ( Beckman-Coulter ) and distributed one cell per well on a 96-well plate , pre-seeded with uninfected Vero cells . To ensure reproducibility between the experiments , each experiment uninfected cells were sorted first , and the average and median fluorescence were kept at ~30 fluorescence arbitrary units ( AU ) . The fluorescence cut-offs for these experiments were set to 100 ( low to intermediate ) and 500 ( intermediate to high ) AU for Vero and gHeLa . HFF expressed lower fluorescence levels , thus the cut-offs for these cells were set to 100 and 350 AU . To verify that only one cell was placed in each well , the 96-well plate was scanned under a fluorescence microscope ( Nikon Eclipse Ti-E ) at 24 and 48 HPI to verify that only one infectious center was observed in each well . When an infected cell was detected at 24 HPI , but did not develop into an infectious center , we determined it as an abortive infection . We conducted three experiments in Vero cells , three experiments in HFF and four in HeLa cells; all experiments were performed in MOI 10 and 100 . For Vero cells , 200 single cells were identified , of which 142 were further analyzed . Ten wells ( ~2% ) contained more than one plaque focus . For HFF , 192 single cells were identified and 141 were further analyzed with qPCR . 19 wells ( ~3% ) contained plaques with at least two foci . For HeLa cells , 124 productive single cells were identified and analyzed . In addition to these cells , 61 infected single cells were identified that did not develop into infectious centers . Fourteen percent of the wells contained at least two focus plaques . Wells with one infectious center were incubated for 6 dpi ( days post infection ) , and then incubated for one hour with lysis buffer containing 10mM Tris-HCl , pH = 8 . 0 , 1mM EDTA ( Merck ) , 1% Tween 20 ( Sigma-Aldrich ) , 0 . 04% Proteinase K ( BIO-LAB , Israel ) [57] . Proteinase K was deactivated by exposure to 95°C for 10 minutes . The lysate was kept at 37°C for an additional 24 hours , prior to analysis with qPCR . The samples were analyzed with qPCR ( QuantStudio 12K Flex , Applied Biosystems ) , using SYBR master mix ( Applied Biosystems ) . A list of primers used for the qPCR assay is specified in S3 Table . Parameters of each analyzed single cell were collected during sorting with Summit 6 . 2 ( Beckman-Coulter ) . These included FSC , SSC and fluorescence ( 561/14 ) . Further analysis measured the number of barcodes for each single cell . The Thompson tau method was used as an initial analysis to discover fluorescence/barcode outliers . Of 142 Vero , 141 HFF and 124 gHeLa single cells that were analyzed , four Vero , three HFF and five gHeLa single cells in MOI 10; and one Vero , three HFF and two gHeLa single cells in MOI 100 were outliers . After their removal , 137 Vero , 135 HFF and 117 gHeLa single cells were used for further analysis . We performed student's t-test and one-way ANOVA statistical tests and calculated Pearson correlation with SPSS statistical software ( IBM ) . To test whether Poisson distribution was a good approximation for number of barcodes per cell distribution , we used the Chi-Square Goodness of fit test . Since only infected cells where examined , we considered distribution of positive numbers , thus excluding the option of zero ( i . e . uninfected cells ) . This required an adjustment of the Poisson distribution with Maximum Likelihood estimate . For testing viral gene expression maintenance throughout the infection , Vero cells were infected with an equal mixture of 14 barcoded recombinants in MOI 10 or 100 . Infections were monitored every 10 minutes for 18 hours . Four frames per well and three wells per MOI were visualized . For testing color change throughout the infection , Vero cells were infected with an equal mixture of 3 color recombinants in MOI 100 . Infections were monitored every 15 minutes for 16 hours . All the movies were taken using a Nikon Eclipse Ti-E epifluorescence inverted microscope with a heated stage top incubator ( Chamlide , live cell instrument ) . The movies were analyzed with Imaris software ( Bitplane ) to discover the increase in mCherry fluorescence . Data were normalized by a moving average of five time points for each cell . We note that for each movie , several images showed a dramatic decrease in the fluorescence level for both channels . This anomaly reflected random electrical power problems . To overcome this , we used an average of time points before and after the decrease as an estimate for fluorescence level . Five different experiments were performed on different days , for each experiment from 600 , 000 to 1 , 150 , 000 cells per population were taken ( population size was equal within each experiment ) . Cells infected with an MOI 10 or 100 were trypsinized at 3 HPI and centrifuged at 1 , 500 rpm ( 264g ) in 4°C for 2 minutes . Cells were resuspended in FACS buffer ( Dulbacco's phosphate buffered saline PBSX1 , 5mM EDTA , 1% FBS ) . Cells were sorted with Astrios ( Beckman-Coulter ) into three populations according to their fluorescence ( lowest , middle , highest ) . After sorting , cells were centrifuged in 1 , 500rpm in 4°C for 10 minutes . After the removal of the supernatant , Bio-Tri RNA Reagent ( BIO-LAB , Israel ) was added according to the number of sorted cells and immediately frozen in -80°C . The RNA was produced following the manufacturer protocol , with materials adjusted to the number of sorted cells and immediately frozen in -80°C . The RNA samples were quantified with Nanodrop ( Thermo Scientific ) and quality of the RNA sample was determined . Equal amounts of RNA ( per experiment ) were taken to produce cDNA . cDNA was produced , using random primers , with RevertAid first strand cDNA kit ( Thermo scientific ) or High-Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) and the resulting cDNA was used for qPCR analysis ( QuantStudio 12K Flex , Applied Biosystems ) , using SYBR master mix ( Applied Biosystems ) . Each cDNA sample was analyzed with viral immediate-early , early , late genes , and a cellular housekeeping gene ( HMBS ) . The sequences for primers for viral genes and cellular housekeeping gene appear in S4 Table .
|
Single cell variation is of major interest in understanding key biological processes , like cancer , development and host pathogen interaction . During viral infection , these cell to cell variations can change the outcome of the whole organism infection . We suggested that differences in the number of parental viral genomes that initiate the replication process alter the outcome of infection among single cells . In this work we present a method based on genetically barcoded herpesvirus recombinants to identify the number of viral genomes initiating replication in individual cells . Our results indicate that viral gene expression is tightly linked to the number of viral genomes replicating per cell . Remarkably , we found that high cellular gene expression was an indicator for a lower viral gene expression in a given cell . We suggest that variations among single cells result from preexisting differences among cells , as well as from random viral host interactions .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"vero",
"cells",
"hela",
"cells",
"biological",
"cultures",
"microbiology",
"light",
"microscopy",
"virus",
"effects",
"on",
"host",
"gene",
"expression",
"microscopy",
"viral",
"genome",
"cell",
"cultures",
"microbial",
"genetics",
"microbial",
"genomics",
"research",
"and",
"analysis",
"methods",
"viral",
"genomics",
"fluorescence",
"microscopy",
"gene",
"expression",
"cell",
"lines",
"viral",
"replication",
"viral",
"genetics",
"virology",
"genetics",
"viral",
"gene",
"expression",
"biology",
"and",
"life",
"sciences",
"cultured",
"tumor",
"cells",
"genomics"
] |
2016
|
Gene Expression Correlates with the Number of Herpes Viral Genomes Initiating Infection in Single Cells
|
Dendritic cells ( DCs ) contribute to human immunodeficiency virus type 1 ( HIV-1 ) transmission and dissemination by capturing and transporting infectious virus from the mucosa to draining lymph nodes , and transferring these virus particles to CD4+ T cells with high efficiency . Toll-like receptor ( TLR ) -induced maturation of DCs enhances their ability to mediate trans-infection of T cells and their ability to migrate from the site of infection . Because TLR-induced maturation can be inhibited by nuclear receptor ( NR ) signaling , we hypothesized that ligand-activated NRs could repress DC-mediated HIV-1 transmission and dissemination . Here , we show that ligands for peroxisome proliferator-activated receptor gamma ( PPARγ ) and liver X receptor ( LXR ) prevented proinflammatory cytokine production by DCs and inhibited DC migration in response to the chemokine CCL21 by preventing the TLR-induced upregulation of CCR7 . Importantly , PPARγ and LXR signaling inhibited both immature and mature DC-mediated trans-infection by preventing the capture of HIV-1 by DCs independent of the viral envelope glycoprotein . PPARγ and LXR signaling induced cholesterol efflux from DCs and led to a decrease in DC-associated cholesterol , which has previously been shown to be required for DC capture of HIV-1 . Finally , both cholesterol repletion and the targeted knockdown of the cholesterol transport protein ATP-binding cassette A1 ( ABCA1 ) restored the ability of NR ligand treated cells to capture HIV-1 and transfer it to T cells . Our results suggest that PPARγ and LXR signaling up-regulate ABCA1-mediated cholesterol efflux from DCs and that this accounts for the decreased ability of DCs to capture HIV-1 . The ability of NR ligands to repress DC mediated trans-infection , inflammation , and DC migration underscores their potential therapeutic value in inhibiting HIV-1 mucosal transmission .
Worldwide , heterosexual transmission accounts for most new HIV-1 infections , with a majority of these occurring in developing countries [1] , [2] . Clearly , controlling heterosexual transmission of HIV-1 would be a significant step toward reducing this global pandemic . To achieve this goal , it will be important to delineate the cellular and molecular events that promote or restrict virus transmission and dissemination . Immune cells within the vaginal , cervical , or rectal mucosa are thought to be the primary targets of infection in the sexual transmission of HIV-1 [1] , [3] , [4] . These target cells include sub-epithelial CD4+ T lymphocytes , intra-epithelial Langerhans cells , macrophages , submucosal plasmacytoid DCs ( pDCs ) , and myeloid ( or conventional ) DCs ( mDCs ) located within the lamina propria [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] . DCs , in particular , play a central role in HIV-1 transmission . DCs are thought to capture cell-free HIV-1 particles from the intralumenal space or from the mucosa after transcytosis across or leakage of HIV-1 particles through the epithelial barrier or by contacting HIV-1-infected cells introduced into the mucosa through abrasions or ulcerative lesions [6] , [12] , [13] . In addition , studies examining vaginal transmission of SIVmac in a rhesus macaque model of AIDS have implicated DCs in virus dissemination from the mucosa to draining lymph nodes [6] , [14] . Moreover , DCs are the predominant infected migratory cell type harboring HIV-1 from virus exposed cervical tissue explants [15] supporting the idea that they are involved in virus dissemination . Upon capture , DCs can deliver infectious HIV-1particles to draining lymph nodes that contain large numbers of CD4+ T cells [16] , [17] . The close contact between virus-laden DCs and CD4+ T cells facilitates cell-to-cell transmission and viral spread [18] , [19] . In addition to their roles in virus transmission and dissemination , DCs can produce proinflammatory cytokines that create a microenvironment that favors virus replication [20] , [21] , [22] . Recent reports have demonstrated that DCs matured by exposure to pathogens encoding Toll-like receptor ( TLR ) ligands or to proinflammatory cytokines are capable of enhanced HIV-1 trans-infection [23] , [24] , [25] and chemokine-directed migration [26] , [27] , suggesting that agents capable of preventing inflammation and DC maturation may be able to limit HIV-1 transmission and dissemination . NRs are a superfamily of ligand-activated transcription factors that includes classic hormone receptors , as well as the so-called orphan receptors and adopted orphan receptors whose natural ligands are either unknown or recently discovered [28] , [29] . Included in these latter two families are peroxisome-proliferator activated receptors ( PPAR ) and liver X receptors ( LXR ) . Ligand-activated PPARγ and LXR are bifunctional modulators of gene expression , capable of either activating or repressing transcription in a promoter-specific manner . Importantly , PPARγ and LXR are potent inhibitors of inflammation and are capable of repressing cytokine and chemokine production by Toll-like receptor ( TLR ) -activated macrophages and DCs through trans-repression mechanisms involving the failure to clear co-repressor complexes from promoters or through direct antagonism of transcription factors such as the p65 subunit of NF-κB , AP-1 , STATs , and IRF3 [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] . The effects of PPARγ and LXR on TLR signaling are complex and a number of studies have demonstrated that each NR inhibits different subsets of inflammatory genes [32] , [34] . For example , LXR signaling represses TLR4-induced expression of iNOS , COX-2 , and IL-6 in murine macrophages , while PPARγ signaling represses IL-1β , GCSF , MCP-1 , MCP-3 , and MIP-1α expression [32] . Here , we show that PPARγ and LXR signaling acutely prevents TLR-activated expression of the proinflammatory cytokines TNF-α , IL-6 , and IL-8 , which have been implicated as co-factors for enhanced mucosal transmission of HIV-1 . Moreover , PPARγ and LXR signaling inhibit the expression of the chemokine receptor CCR7 , thereby preventing DC chemotaxis in response to gradients of CCL21 , a process thought to be involved in DC migration from mucosal surfaces to draining lymph nodes . As opposed to their inhibitory effects on inflammatory gene expression , ligand-activated PPARγ and LXR induce expression of genes involved in lipid and cholesterol metabolism , as well as cholesterol transport , including ABCA1 and ABCG1 [29] , [39] , [40] , [41] . Importantly , many studies have demonstrated that cholesterol plays an essential role in HIV-1 biology . Cholesterol must be present in both the target cell membranes and HIV-1 particles for efficient virus binding and fusion [42] , [43] , [44] , [45] , [46] . In addition , nascent HIV-1 particles bud through cholesterol-rich lipid rafts [47] , [48] and infectious particles enter target cells through cholesterol-rich lipid rafts [42] , [49] , [50] . Finally , studies using the cholesterol chelator , methyl-β-cyclodextrin , demonstrated that cholesterol is required for DC binding of virus particles [51] . Interestingly , PPARs and LXR are expressed at high levels in HIV-1 target cells such as macrophages and DCs [28] , [29] . Therefore , we hypothesized that PPARγ and LXR-mediated changes in cholesterol metabolism and trafficking might contribute to their ability to inhibit the transmission of HIV-1 from DCs to T cells . Our results demonstrate that PPARγ and LXR signaling inhibit the capture of HIV-1 by DCs , and its subsequent transfer to CD4+ T cells . These effects are due to up-regulation of ABCA1-dependent cholesterol efflux , a mechanism distinct from the effects of PPARγ and LXR signaling on DC migration and proinflammatory cytokine production . Collectively , our data suggest that the bifunctional activities of ligand activated PPARγ and LXR can be exploited to inhibit multiple distinct steps in HIV-1 mucosal transmission and dissemination .
TLR signaling induced by sexually transmitted pathogens is thought to enhance HIV-1 mucosal transmission in part by promoting local inflammation . Inflammation not only activates HIV-1 target cells but , importantly , it also induces DC maturation and the subsequent migration of HIV-1-carrying DCs to local lymph nodes where they can contribute to virus dissemination [16] , [17] . We were therefore interested in determining whether the anti-inflammatory activities of ligand-activated PPARγ and LXR [34] , [52] , [53] could be exploited to limit DC functions involved in HIV-1 transmission and pathogenesis . To examine the effects of PPARγ and LXR signaling on DC maturation , human monocyte-derived DCs ( MDDCs ) were treated with E . coli K12 LPS , a TLR4 ligand , in the presence or absence of ligands for PPARγ and LXR . As expected , LPS treatment upregulated the expression of surface markers associated with maturation , such as HLA-DR , CD80 , CD86 , and CD83 , downregulated the expression of surface markers associated with an immature phenotype , such as the C-type lectin DC-SIGN , but had no effect on the expression of the pan-DC marker CD11c ( Figure 1A and data not shown ) . Notably , treatment of MDDCs with the PPARγ ligand ciglitazone or the LXR ligand TO-901317 inhibited LPS-dependent upregulation of cell-surface expression of HLA-DR , CD80 , and CD86 ( Figure 1A ) . Similarly , we found that ciglitazone or TO-901317 treatment inhibited human MDDC maturation in response to the TLR2 ligand PAM3CSK4 ( data not shown ) . We next examined the effects of ciglitazone and TO-901317 treatment on TLR-induced proinflammatory cytokine and chemokine production . We found that treatment with these PPARγ and LXR ligands prevented the release of proinflammatory cytokines and chemokines such as TNF-α , IL-6 , and IL-8 by PAM3CSK4-activated MDDCs ( Figure 1B ) . In addition , PPARγ and LXR treatment also prevented the release of the chemokines MIP-1α and RANTES , which are important for the recruitment of CD4+ T cells to sites of infection , both from MDDC in response to the TLR4 ligand LPS ( Figure 1C ) and from plasmacytoid DCs ( pDCs ) in response to the TLR7 ligand CLO97 and the TLR9 ligand CpG ODN 2006 ( Figure 1D ) . Importantly , PPARγ and LXR signaling inhibited TLR-induced proinflammatory cytokine and chemokine production coincident with TLR ligation ( data not shown ) , suggesting that NR-mediated inhibition most likely acts through a trans-repression mechanism [34] . The concentrations of the PPARγ ligand ciglitazone and the LXR ligand TO-901317 necessary to see a reduction in DC maturation and the production of pro-inflammatory cytokines and chemokines did not affect MDDC viability as measured by LDH release or mitochondrial activity ( Figure S1 and data not shown ) . In addition to transmitting HIV-1 to T cells with high efficiency , DCs can also contribute to HIV-1 pathogenesis by binding virus and then migrating from mucosal sites of infection to regional lymph nodes . In this way , DCs can contribute to viral dissemination . Studies have shown that mature DCs have a greater migratory capacity than immature DCs [26] , [27] . This led us to examine whether NR signaling would also inhibit MDDC migration through a 5 µm pore size Transwell insert in response to the chemokine CCL21 , which has been shown to be important for DC migration in vivo [27] . We found that LPS-matured MDDCs ( mMDDCs ) migrated in response to a CCL21 gradient and that co-treatment with PPARγ or LXR ligands repressed this migration approximately 2-fold ( Figure 2A ) . In contrast , immature MDDCs ( iMDDCs ) migrated quite poorly in response to CCL21 and , consequently , NR ligand treatment had a limited effect . Expression of CCR7 , a receptor for CCL21 , is upregulated in DCs in response to TLR engagement [26] , [54] . Notably , treatment with PPARγ and LXR ligands prevented the LPS-induced upregulation of CCR7 ( Figure 2B ) , which may partly explain why NR ligand-treated MDDCs migrate poorly in response to CCL21 . Together , these data suggest that PPARγ and LXR signaling inhibit DC migration by preventing TLR-induced DC maturation . DCs are thought to play a critical role in virus dissemination by capturing HIV-1 and transferring it to T cells [5] , [24] , [55] . We therefore examined whether NR ligands could modulate DC-mediated HIV-1 trans-infection . iMDDCs were treated with ciglitazone or TO-901317 for 48 hours , extensively washed , and then incubated for four hours with either a single-round replication-competent HIV-1 reporter virus packaged with an R5-tropic envelope or with wild-type HIV-1 . Following incubation with the virus , MDDCs were washed extensively to remove unbound virus and then cultured directly with autologous T cells or in the upper well of a Transwell insert separated from the T cells by a 0 . 4 µm membrane . Although HIV-1 replicated very poorly in immature MDDCs ( Figure 3 ) , we found that DCs were able to mediate T cell infection when directly cultured with the T cells or when separated from them by the Transwell insert ( Figure 3 ) , suggesting that a portion of the MDDC-mediated trans-infection is mediated by either exosome-associated HIV-1 [56] or virus shed from the surface of MDDCs [57] . Most importantly , we found that PPARγ and LXR ligands inhibited trans-infection up to 5-fold underscoring their potential to limit HIV-1 transmission ( Figure 3 ) . NR signaling inhibits trans-infection of T cells by both single-round replication competent virus ( Figure 3 ) and wild-type replication competent virus ( Figure 4A ) , suggesting that the majority of virus transferred to T cells is due to virus captured by the DC and not due to newly synthesized virus . Because mature DCs capture and transfer HIV-1 to T cells with higher efficiency than immature DCs [23] , [24] , [25] , we next determined whether PPARγ or LXR ligands could inhibit trans-infection mediated by LPS- or PAM3CSK4-matured MDDCs . PPARγ and LXR signaling repressed trans-infection of autologous primary T cells mediated by both immature , LPS-matured MDDCs ( Figure 4A ) , and PAM3CSK4-matured MDDCs ( Figure 4B ) , suggesting that the repression is independent of MDDC maturation . To confirm NR-dependent maturation-independent repression of DC-mediated HIV-1 trans-infection , we matured MDDCs with LPS for two days prior to treatment with NR ligands and then assayed for HIV-1 transfer . As seen in figure 4C , the ability of mature MDDCs to transfer virus was impaired when treated with PPARγ and LXR ligands . In addition , we found that PPARγ and LXR ligand treatment of MDDCs prevented trans-infection over a wide range of input virus ( Figure 4D ) . Of note , NR-ligand treatment inhibited immature and mature MDDC-mediated trans-infection of both R5- and X4-tropic envelope glycoprotein-pseudotyped single-round replication competent reporter viruses and replication-competent R5- and X4- tropic wild-type HIV-1 ( data not shown ) . Together these data suggest that , unlike PPARγ- and LXR-mediated inhibition of migration , the inhibition of trans-infection is independent of the maturation state of the DC . Importantly , MDDC-mediated trans-infection is also inhibited by rosiglitazone ( Figure 4E ) , a PPARγ agonist that is currently licensed for the systemic treatment of type II diabetes . Next , we wanted to examine the mechanism accounting for the inhibition of trans-infection . We began by examining the effects of PPARγ or LXR ligand treatment on HIV-1 binding to MDDCs . Ciglitazone and TO-901317 treatment led to a 2 to 5-fold decrease in the amount of HIV-1 associated with MDDCs as measured by an ELISA for the HIV-1 p24 capsid protein ( Figure 5A ) . Another PPARγ ligand , rosiglitazone , was also tested and had a comparable effect on HIV-1 capture ( Figure 5B ) . Treatment with these NR ligands also inhibited the capture of HIV-1 by DCs at 4°C , suggesting that NR ligand treatment prevents DC binding of HIV-1 ( Figure S2 ) . In addition , we found that PPARγ and LXR ligand treatment of MDDCs prevented capture over a wide range of input virus ( Figure 5C ) . Although NR signaling can repress inflammatory gene expression by a trans-repression mechanism [30] , [31] , [32] , [33] , [34] , [36] , [37] , [38] , [52] , it likely decreases HIV-1 capture through a different mechanism . MDDCs must be treated with PPARγ and LXR ligands for at least 12 hours in order to observe inhibition of virus capture ( Figure 5D ) , suggesting that changes in cellular gene expression are required for the observed effect . Though the amount of virus captured by MDDCs upon NR ligand treatment was reduced , the relative amount of virus particles internalized was similar ( Figure 5E ) suggesting that reduced ability of MDDCs to capture HIV-1 particles upon NR ligand treatment was not due to gross reduction in cellular endocytic function . To confirm that NR ligand treatment does not alter the ability of MDDCs to internalize particles , we examined their effects on the ability of MDDCs to macropinocytose FITC-labeled dextran . NR ligand treatment had no effect on FITC-dextran internalization by immature or mature MDDCs ( Figure S3 and data not shown ) . Our data suggest that changes in cellular gene expression are necessary for the observed decrease in HIV-1 capture by MDDCs . We therefore considered the possibility that PPARγ and LXR ligand treatment altered the expression of known HIV-1 attachment factors expressed on the surface of immature MDDCs . However , we found that NR ligand treatment did not alter the expression of CD4 , CCR5 , or DC-SIGN ( Figure 5F ) , which have been implicated in DC capture of HIV-1 [5] , [58] . Despite these findings , we cannot rule out whether NR signaling alters the expression of other factors implicated in HIV-1 attachment such as other C-type lectins [59] , [60] , [61] , [62] , heparan sulfate proteoglycans [63] , [64] , [65] , or GSLs [66] , [67] , [68] , [69] , [70] . Although MDDCs are a faithful representation of myeloid or conventional DCs ( mDCs ) with respect to their interactions with HIV-1 [25] , we decided to utilize mDCs freshly isolated from the peripheral blood of healthy volunteers . We found that PPARγ and LXR signaling inhibited the ability of immature and LPS-matured mDCs to capture HIV-1ADA and transfer it to autologous T cells ( Figure 6 ) in a manner consistent with results obtained using MDDCs . Because direct DC-T cell contact is required for efficient virus transfer [24] , [57] ( and Figure 3 ) , we wanted to determine whether NR ligand treatment interfered with the ability of MDDCs to form conjugates with T cells . Using a FACS-based conjugate formation assay [71] , we determined that NR ligand-treated MDDCs were able to form conjugates with primary autologous T cells in a manner similar to untreated MDDCs ( Figure 7A ) . Because NR ligand treatment did not alter the ability of DCs to form conjugates with T cells , we next wanted to examine whether such treatment prevented the formation of functional virological synapses between DCs and T cells . Confocal microscopy data suggest that PPARγ and LXR ligand-treated DCs are capable of forming virological synapses , as indicated by co-localization of virus and the tetraspanin CD81 at the site of DC-T cell contact ( Figure 7B ) . However , number of virus particles localized at the virological synapse is decreased in NR ligand-treated cells . Taken together , our data suggest that NR signaling impairs the ability of MDDCs to transfer virus to T cells by inhibiting the capture of HIV-1 by MDDCs . Recent studies have demonstrated that DCs can bind to infectious HIV-1 and envelope-deficient virus-like particles ( VLPs ) in a GSL-dependent , viral envelope glycoprotein-independent manner [72] , [73] . We therefore wanted to assess whether triggering PPARγ and LXR signaling could alter the ability of MDDCs to bind virus independently of the envelope glycoprotein gp120 . We found that PPARγ and LXR ligand treatment led to a 2 to 5-fold decrease in the amount of envelope glycoprotein ( Env ) -deficient HIV-1 particles captured by both immature and mature MDDCs ( Figure 8A ) , suggesting that GSL-based virus-DC interactions may be targeted by NR signaling . To demonstrate that this interaction is truly envelope glycoprotein-independent , we also examined the effects of PPARγ and LXR signaling on the ability of DCs to capture HIV-1 particles pseudotyped with the glycoproteins of vesicular stomatitis virus ( VSV ) , Ebola virus ( EboV ) , and Marburg virus ( MarV ) . As shown in figure 8B , treatment with PPARγ and LXR ligands inhibited the ability of DCs to capture EboV or MarV glycoprotein-pseudotyped HIV-1 particles , whereas the treatment had no effect on the ability of DCs to capture VSV-G-pseudotyped particles . Since , like HIV-1 , both EboV and MarV glycoproteins are known to require cholesterol for infection [74] , [75] , whereas VSV-G does not [43] , [75] , [76] , this suggested that PPARγ and LXR might be exerting their effects through the regulation of cellular cholesterol . Previous studies have shown that DC capture of HIV-1 is dependent upon the cholesterol content of the cell membrane [51] . Since both PPARγ and LXR are known to modulate genes involved in cholesterol metabolism and transport [29] , [39] , [40] , [77] , we were interested in determining whether ciglitazone or TO-901317 affected the cholesterol content of MDDCs . Treatment with PPARγ and LXR ligands increased cholesterol efflux from immature MDDCs approximately 2 to 3-fold ( Figure 9A ) and led to a concomitant 2-fold decrease in the amount of cholesterol in immature MDDCs ( Figure 9B ) . We next wanted to see if cholesterol depletion resulting from PPARγ and LXR ligand treatment was responsible for the decreased ability of MDDCs to capture and transfer HIV-1 . In order to do this , we replenished membrane cholesterol in PPARγ and LXR ligand-treated MDDCs using cholesterol-saturated methyl-β-cyclodextrin and assayed for HIV-1 capture and transfer to T cells . Cholesterol repletion of NR ligand-treated MDDCs with cholesterol-saturated methyl-β-cyclodextrin restored cholesterol content ( Figure 9B ) and , importantly , fully restored the ability of both immature and mature MDDCs to capture HIV-1 ( Figure 9C ) and transfer it to CD4+ T cells ( Figure 9D ) . PPARγ and LXR signaling upregulate expression of ATP-binding cassette protein A1 ( ABCA1 ) that facilitates the apoA1-dependent efflux of cholesterol from cells [39] , [41] , [77] . We therefore examined whether treatment of DCs with ciglitazone or TO-901317 affected ABCA1 expression . We found by western blot analysis that both NR ligands increased ABCA1 expression ( Figure 9E ) . Importantly , targeted knockdown of ABCA1 using shRNA abrogated the effect of PPARγ and LXR ligand treatment on cholesterol efflux ( data not shown ) , HIV-1 capture by DCs ( Figure 9F ) , and HIV-1 transfer to T cells ( Figure 9G ) . These findings suggest that ligand-activated PPARγ and LXR mediate their effects through the depletion of cholesterol from the DC plasma membrane via the up-regulation of the ABCA1 cholesterol transport protein . It will be interesting to determine whether HIV-1 particles interact directly with cholesterol in the plasma membrane of DCs or with factors that localize to cholesterol-rich lipid rafts .
Sexual transmission of HIV-1 is enhanced by inflammatory and ulcerative co-infections with STI pathogens that cause diseases such as genital herpes , gonorrhea , syphilis , Chlamydia , bacterial vaginosis , and fungal infections [78] , [79] , [80] , [81] , [82] , [83] . This enhanced susceptibility to infection may be due to a number of factors , including disruption of epithelial integrity [6] , [11] , [14] , [84] , [85] , [86] , [87] , recruitment of HIV-1 target cells such as Langerhans cells , DCs , macrophages , and T lymphocytes to sites of inflammation [8] , and activation of HIV-1 expression by pro-inflammatory cytokines or microbial components [21] , [22] , [88] , [89] , [90] , [91] , [92] . It is likely that STI pathogens enhance these latter two processes , at least in part , through engagement of the TLR family of innate immune receptors . Clearly , prophylactic methods that inhibit infection of the genital or rectal mucosa would significantly limit the global spread of HIV . To this end , considerable efforts have been directed toward the development of microbicides that interfere with virus integrity or with key steps in virus replication . However , to date , little attention has been paid to targeting cellular pathways involved in active suppression of inflammation and its effects on mucosal HIV-1 infection and virus dissemination . With this in mind , we have focused our efforts on the nuclear receptor family of transcription factors that have recently been shown to be potent inhibitors of TLR-induced inflammation [30] , [31] , [32] , [33] , [34] , [36] , [37] , [38] . Here , we demonstrate that PPARγ and LXR signaling inhibit several aspects of DC biology that are important for HIV-1 mucosal transmission . These include TLR-induced pro-inflammatory cytokine expression , DC migration in response to the chemokine , CCL21 , and , importantly , DC-mediated capture of infectious virus particles and trans-infection of CD4+ T cells . Our findings highlight the therapeutic potential of PPARγ and LXR ligands as topical treatments that could be used in conjunction with conventional microbicides to limit mucosal transmission of HIV-1 . DC-mediated trans-infection of T cells is thought to play a critical role in the mucosal transmission of HIV-1 . Studies suggest that DCs can mediate trans-infection either by internalizing infectious virions into a protected tetraspanin-rich intracellular compartment , or deep membrane invaginations contiguous with the cell surface , and releasing them for the subsequent infection of T cells [5] , [56] , [73] , [93] , [94] , [95] or by retaining virions at the cell surface and transferring them to T cells [57] , [95] . Regardless of the mechanism , maturation of DCs with ligands for TLRs such as TLR4 and TLR2/TLR1 increases DC-mediated HIV-1 capture and trans-infection of T cells . DC maturation also contributes to HIV-1 mucosal transmission in a number of other ways . Mature DCs create a pro-inflammatory environment that favors virus replication [20] , [88] , [96] , [97] , [98] and leads to disruption of the mucosal integrity [83] , [99] . Mature DCs may also contribute to virus dissemination by virtue of their enhanced ability to traffic to regional lymph nodes in response to chemokine gradients and , once there , transfer virus to resident CD4+ T cells . Here we show that PPARγ or LXR ligand treatment can prevent DC maturation as measured by the expression of cell surface markers such as HLA-DR , CD80 , and CD86 ( Figure 1A ) . Importantly , treatment with PPARγ or LXR ligands also potently inhibit expression of maturation-associated pro-inflammatory cytokines ( Figure 1B ) , such as TNF-α and IL-6 and the pro-inflammatory chemokine , IL-8 , that have been shown to augment HIV-1 replication in infected cells and to increase HIV-1 transmission to T cells [21] , [22] , [91] , [100] . Moreover , we demonstrate that PPARγ and LXR signaling can interfere with the migration of DCs in response to a CCL21 chemokine gradient ( Figure 2A ) . This appears to be due to the effects of PPARγ and LXR signaling on the expression of CCR7 ( Figure 2B ) , one of the receptors for CCL21 . CCR7 is up-regulated upon DC maturation and has been shown to be important for the migration of DCs from the mucosa to regional lymph nodes in vivo [54] . By preventing DC migration in response to CCL21 , PPARγ and LXR ligands may help to block the dissemination of DC-associated virus from mucosal sites of infection to regional lymph nodes . Recent studies demonstrated that activation/maturation of DCs through TLR4 or TLR2/TLR1 enhances HIV-1 transmission to target cells via increased HIV-1 capture [23] , [24] , [25] , [92] and Figure 4 and 5 ) . Here , we demonstrate that activating PPARγ or LXR signaling pathways in DCs decreases the ability of both immature and TLR-matured DCs to capture and transfer HIV-1 to T cells ( Figure 3 , 4A and 5A ) . Furthermore , NR signaling can inhibit HIV-1 transfer by previously matured DCs ( Figure 4C ) These results suggest that PPARγ and LXR signaling alter other pathways involved with HIV-1 trans-infection that are independent of the maturation state of the DC ( Figure 4C ) , however we cannot rule out the possibility that the prevention of DC maturation may contribute to the NR-mediated decrease in HIV-1 capture and transfer . Many studies have demonstrated a role for PPARγ and LXR signaling in cholesterol metabolism and transport [29] , [39] , [40] . For example , both signaling pathways stimulate the expression of ABCA1 and ABCG1 , which have been implicated in apolipoprotein A1 ( ApoA1 ) - and high density lipoprotein ( HDL ) -mediated cholesterol efflux , respectively [39] . Given the importance of cholesterol for a number of aspects of HIV-1 biology , including virus binding and infection [42] , [43] , [44] , [45] , [47] , [48] , [49] , [50] , [51] , [76] , we hypothesized that PPARγ and LXR signaling was altering the cholesterol content of DC membranes , thereby rendering them incapable of efficiently binding HIV-1 particles . Previous studies have demonstrated that treatment with cholesterol depleting drugs , such as methyl-β-cyclodextrin , or with cholesterol synthesis inhibitors , such as HMGCoA-reductase inhibitors ( statins ) , alters the ability of cells , including DCs , to bind HIV-1 and renders them refractory to HIV-1 infection [42] , [43] , [45] , [49] , [50] , [51] , [101] . Here , we show that cholesterol repletion of PPARγ and LXR ligand-treated DCs reverses the effects of the NR ligands on virus capture and transfer ( Figure 9C and 9D ) , confirming that PPARγ and LXR are mediating their effects through membrane cholesterol . In addition , targeted shRNA knock-down of ABCA1 abrogates the effects of PPARγ and LXR signaling on HIV-1 capture and transfer ( Figure9F and 9G ) . A recent study suggests that LXR-dependent cholesterol efflux in macrophages is mediated entirely through ABCA1 , with little to no contribution from ABCG1 [102] . We cannot , however , formally exclude a contribution from ABCG1-dependent cholesterol efflux to the effects we report here . Our data show that PPARγ and LXR signaling decrease cellular cholesterol content , which may in turn deplete cholesterol from membrane lipid rafts . It will be interesting to determine whether treatment of DCs with PPARγ and LXR ligands disrupts lipid rafts and whether this accounts for the decreased ability or NR-treated DCs to capture and transfer HIV-1 . We found that PPARγ and LXR ligand treatments do not alter the levels of a number of known virus attachment factors expressed on DCs including CD4 , CCR5 , and DC-SIGN ( Figure 5F ) . Moreover , PPARγ and LXR signaling prevents the capture of Env-deficient HIV-1 virus like particles ( Figure 8A ) , suggesting that virus envelope glycoprotein/receptor interactions are not involved in the observed effect . That the effects of PPARγ and LXR signaling on HIV-1 capture are virus envelope glycoprotein-independent is supported by our finding that treatment of DCs with ciglitazone or TO-901317 prevents the capture of viral particles pseudotyped with the envelope glycoproteins of Ebola virus and Marburg virus ( Figure 8B ) . Interestingly , these two viruses are known to require cholesterol for infection [74] , [75] . In contrast , treatment with the NR ligands had no effect on the ability of DCs to capture virus particles pseudotyped with the envelope glycoprotein of VSV . Previous studies have demonstrated that VSV-G-pseudotyped HIV-1 particles are efficiently captured by cells depleted of cholesterol using methyl-β-cyclodextrin [43] , [75] , [76] . These data further support the hypothesis that PPARγ and LXR signaling alter the membrane cholesterol content of DCs , rendering them incapable of efficiently capturing HIV-1 particles . Although NR ligand treatment limits the expression of immune-activating cytokines and co-stimulatory molecules that are up-regulated as DCs mature , we found that it does not alter the ability of DCs to form conjugates with T cells . The number of DC-T cell conjugates formed with PPARγ and LXR ligand-treated DCs was comparable to that of control untreated DCs ( Figure 7A ) . It will be interesting to determine whether these conjugates represent functional immunological synapses between DCs and T cells . It is worth noting that DC-to-T cell transfer of HIV-1 most likely occurs through the formation of virological synapses [103] , [104] , [105] , [106] , [107] , [108] . We found that NR ligand treatment does not prevent the formation of virological synapses between DCs and T cells as assessed by confocal microscopy , although ligand treatment does seem to decrease the amount of virus concentrated at the virological synapse ( Figure 7B ) . Beyond demonstrating the ability of PPARγ and LXR signaling pathways to prevent DC capture and transfer of virus , our results provide support for a number of observations regarding the interactions between DCs and HIV-1 . First , we demonstrate that immature DCs can transfer single round replication competent virus to T cells through a Transwell insert that prevents direct contact between the two cell types ( Figure 3 ) . Although direct cell-cell contact is required for efficient virus transfer , our data suggest that approximately 20% of infectious virus can be transferred by immature DCs via exosomes or shedding from the cell surface . In contrast , although mature DCs bind approximately 10-fold more virus , less than 10% of transfer is mediated through cell-surface bound viral particles ( Figure 4A ) . Second , our data suggest that a large percentage of virions captured by DCs is internalized or otherwise protected from proteases ( Figure 5E ) . Previous studies have demonstrated that DCs internalize HIV-1 , resulting in either degradation of virus particles [56] , [65] , [109] , establishment of productive infection [110] , [111] , [112] , or sequestration into protected intracellular compartments [56] , [73] , [94] , [95] . Although PPARγ and LXR signaling alters the amount of virus captured by DCs , it does not seem to alter the percentage of captured virus that is internalized by DCs ( Figure 5E ) . This is not surprising , since PPARγ and LXR ligand treatment does not alter the endocytic capacity of DCs , as measured by the internalization of FITC-dextran ( Figure S2 ) . Finally , our data confirm that DCs can bind virus particles in a gp120-independent manner ( Figure 8A ) . Recent reports demonstrate that host cell-derived GSLs incorporated into the budding virus particle play a critical role in mediating HIV-1 capture by immature and mature DCs in a gp120-independent manner [72] , [73] . Taken together with current and previous findings that cholesterol depletion from DC membranes prevents HIV-1 binding [51] ( and Figure 9C ) , these data argue for the presence of a yet-to-be-identified GSL-recognizing attachment factor ( s ) within lipid raft-like membrane microdomains at the surface of DCs whose function is compromised upon NR ligand treatment . NR signaling may have beneficial effects on the prevention of HIV-1 transmission beyond the effects on pro-inflammatory cytokine production , migration , and virus capture and transfer . STIs , through engagement of TLRs , and STI/TLR-induced inflammation , can directly activate HIV-1 replication in infected cells . Our data suggest that both PPARγ and LXR ligands repress HIV-1 replication in DCs ( Figure 3 ) , although the levels of replication in this cell type are quite low . This finding is consistent with previous studies that have shown that PPARγ ligands repress HIV-1 expression in infected monocytes and macrophages [113] , [114] , [115] . Recent findings from our laboratory suggest that NR-mediated repression of HIV-1 replication is due to trans-repression ( T . Hanley and G . Viglianti , manuscript in preparation ) , as is thought to be the case for NR-mediated repression of pro-inflammatory cytokine production [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] . Although our data suggest that the majority of virus transferred to T cells is due to virus captured by DCs , and not due to virus newly synthesized in infected DCs , NR-mediated inhibition of HIV-1 replication may contribute to the inhibition of trans-infection that we report here . By preventing HIV-1 replication , in addition to DC migration , pro-inflammatory cytokine and chemokine production , and trans-infection , PPARγ and LXR ligands may block the dissemination of DC-associated virus from the local site of infection to regional lymph nodes . In the absence of an effective vaccine for HIV-1 , the development of topical microbicides that block the early steps of HIV-1 infection and transmission may represent the best option for containing the spread of this global pandemic . To date , there has been limited success with antiviral microbicides . In order to ensure success with future microbicide development , a much greater understanding of the mechanisms involved in the very early stages of mucosal infection and transmission of HIV-1 , and the role of DCs in HIV-1 pathogenesis , in particular , are required . Our results contribute to a better delineation of the mechanisms underlying the HIV-1 trans-infection activity of DCs , while having implications for the development of new anti-HIV microbicide strategies . PPARγ and LXR ligands are small lipophilic molecules that readily diffuse across cell membranes and might be amenable to topical formulations . Two PPARγ agonists , rosiglitazone and pioglitazone , are currently approved for the systemic treatment of type II diabetes . A limitation of the present study is that we have not yet examined the effects of NR signaling on HIV-1 transmission in the context of a complex tissue model or an animal model . Despite this limitation , the anti-inflammatory and anti-HIV-1 activities of PPARγ and LXR provide a solid rationale for considering them as drug targets that can act synergistically with conventional anti-viral microbicides that target other aspects of mucosal transmission including virion structure , virus binding/entry , or reverse transcription .
This research has been determined to be exempt by the Institutional Review Board of the Boston University Medical Center since it does not meet the definition of human subjects research . Primary human CD14+ monocytes were isolated from the peripheral blood mononuclear cells ( PBMCs ) of healthy donors using anti-CD14 magnetic beads ( Miltenyi Biotec ) per the manufacturer's instructions . CD14+ monocytes ( 1 . 5×106 cells/ml ) were cultured in RPMI 1640 supplemented with 10% FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , 0 . 29 mg/ml L-glutamine , 1000 U/ml IL-4 ( PeproTech ) , and 1400 U/ml GM-CSF ( PeproTech ) for 6-8 days at the end of which the cells acquired an immature dendritic cell phenotype as assessed by flow cytometry ( CD11c+ , DC-SIGN+ , HLA-DRlo , CD80− , CD86− ) . Cells were given fresh medium supplemented with IL-4 and GM-CSF every 2 days . Mature dendritic cells were obtained following 48 hour exposure to 100 ng/ml ultra-pure E . coli K12 LPS or 100 ng/ml PAM3CSK4 . Primary human myeloid DCs ( mDCs ) and plasmacytoid DCs ( pDCs ) were isolated from monocyte- and B cell-depleted PBMCs using anti-CD11c and anti-BDCA4 magnetic beads ( Miltenyi Biotec ) per the manufacturer's instructions . mDCs were cultured in RPMI 1640 with 10% FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , 0 . 29 mg/ml L-glutamine , 1000 U/ml IL-4 , and 1400 U/ml GM-CSF . pDCs were cultured in RPMI 1640 supplemented with 10% FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , 0 . 29 mg/ml L-glutamine , and 10 ng/ml IL-3 ( PeproTech ) . Primary human CD4+ T cells were isolated from CD14-depleted peripheral blood mononuclear cells using anti-CD4 magnetic beads ( Miltenyi Biotec ) per the manufacturer's instructions . CD4+ T cells ( 2×106 cells/ml ) were cultured in RPMI 1640 supplemented with 10% FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , 0 . 29 mg/ml L-glutamine , 50 U/ml IL-2 ( R&D Systems ) , and 5 µg/ml PHA-P ( Sigma ) for 6-8 days at the end of which the cells acquired a memory T cell phenotype as assessed by flow cytometry ( CD3+ , CD4+ , CD45RO+ , CD45RA– ) . 293T cells were cultured in DMEM supplemented with 10% FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , and 0 . 29 mg/ml L-glutamine . MAGI-CCR5 cells were cultured in DMEM supplemented with 10% FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , 0 . 29 mg/ml L-glutamine , 500 µg/ml G418 , 1 µg/ml puromycin , and 0 . 1 µg/ml hygromycin B . PM1 cells were cultured in RPMI 1640 supplemented with 10% FBS , 100 U/ml penicillin , 100 µg/ml streptomycin , and 0 . 29 mg/ml L-glutamine . The LXR ligand TO-901317 was purchased from Calbiochem . The PPARγ ligands ciglitazone and rosiglitazone were purchased from Cayman Chemicals . The ligands were reconstituted in DMSO . The TLR2 ligand PAM3CSK4 , the TLR4 ligand E . coli K12 LPS , the TLR7 ligand CLO97 , and the TLR9 ligand CpG ODN 2006 were purchased from Invivogen . Unless otherwise noted , DCs were treated with PPARγ and LXR ligands for 24–48 hours , beginning one hour prior to treatment with TLR ligands . Replication competent R5-tropic HIV-1ADA and X4-tropic HIV-1NL4-3 were generated by infection of PM1 cells . Single-round replication-competent HIV-1-based reporter viruses were generated by packaging a luciferase expressing reporter virus , BruΔEnvLuc2 , with the envelope glycoproteins from CCR5-tropic HIV-1 ( Ada-M ) , CXCR4-tropic HIV-1 ( HXB2 ) , VSV ( VSV-G ) , Ebola virus Zaire ( EboV-Z ) , Ebola virus Sudan ( EboV-S ) , or Marburg virus ( MarV ) . EGFP-labeled virus particles were generated by co-transfection of the pro-viral clone HIV-1NL4-3 with an expression vector encoding a Vpr-EGFP fusion protein . Virus stocks were generated by transfecting HEK293T cells using the calcium phosphate method . All viruses were titered on MAGI-CCR5 cells and p24gag content was determined by ELISA . To assess DC-mediated transfer of HIV-1 to T cells , DCs were incubated with Ada-M- or HXB2-pseudotyped HIV-luciferase reporter virus at an MOI = 0 . 1 ( 37 . 8–40 . 4 ng p24gag ) for four hours at 37°C . Cells were washed five times with PBS to remove unbound virus , seeded in 96-well plates ( 2 . 5×105 cells/well ) , and then cultured with either PM1 T cells ( 5×105 cells/well ) or autologous primary CD4+ T cells ( 5×105 cells/well ) for 48 hours . In some instances , the DCs were seeded in 24-well plates separated from the T cells by a Transwell insert ( Corning ) with a 0 . 4 µm pore size . As controls , virus-exposed DCs and virus-exposed T cells were cultured alone for 48 hours . After 48 hours , the cells were harvested , washed two times with PBS , and lysed in PBS/0 . 02% Triton X-100 . Protein levels in cell lysates were determined using a modified Lowry protein assay ( BioRad ) and luciferase activity was measured using luciferase reagent ( Promega ) and a MSII luminometer ( Molecular Devices ) . In some experiments , replication competent R5-tropic HIV-1ADA or X4-tropic HIV-1NL4-3 ( 5 ng p24gag ) were used in place of packaged reporter virus and transfer was measured by p24gag ELISA . DCs ( 2 . 5×105 cells/well ) were incubated with replication competent R5-tropic HIV-1ADA or X4-tropic HIV-1NL4-3 ( 5 ng p24gag ) for three to four hours at 37°C . Cells were washed four to five times with PBS to remove unbound virus , and lysed in PBS/10%FBS/0 . 5% Triton X-100 . In some experiments , virus-exposed DCs were incubated with 0 . 5% trypsin for 5 minutes at 37°C to degrade surface-bound virus particles , washed twice in culture medium , and then lysed as above . An ELISA was used to determine the amount of p24gag protein associated with the cells . In some experiments , Ada-M- , HXB2- , VSV-G , EboV- , or MarV-packaged HIV-luciferase reporter virus ( 5 ng p24gag ) or an equal amount of reporter virus lacking envelope glycoproteins ( ΔEnv ) was used . Primary CD4+ T cells were labeled with the cytoplasmic dye CMTMR ( CellTracker Orange , Molecular Probes ) for 30 minutes at 37°C and then washed three times with PBS to remove excess dye . Cells were then incubated for 16 hours at 37°C and washed twice with PBS prior to use in conjugate formation assays . Following labeling , 5×105 T cells were incubated with 2 . 5×105 unlabeled iMDDCs for four hours at 37°C in a total volume of 200 µl . The conjugates were then fixed by gently adding an equal volume of 4% paraformaldehyde . Samples were run immediately through a flow cytometer . Conjugate formation was assessed by fluorescence associated with the MDDC population . Primary CD4+ T cells were labeled with the cytoplasmic dye CMCA ( CellTracker Blue , Molecular Probes ) for 30 minutes at 37°C and then washed three times with PBS to remove excess dye . T cells were then incubated for 16 hours at 37°C and washed twice with PBS prior to use in virological synapse formation assays . 2 . 5×105 unlabeled mMDDCs were incubated with 100 ng HIV-1NL4-3 virions packaged with Vpr-EGFP for four hours at 37°C , washed four times with PBS , and incubated with 5×105 CMCA-labeled autologous T cells for four additional hours . The cells were fixed in 1% paraformaldehyde , stained with anti-CD81-PE ( BD Pharmingen ) . Z-stacks were captured on the Nikon deconvolution wide-field Epifluorescence Scope at 100× . Using ImageJ software , the images were deconvolved and the fluorescence was summed . Cholesterol-saturated methyl-β-cyclodextrin was prepared as previously described [116] . Briefly , cholesterol powder was added to 240 mM methyl-β-cyclodextrin solution at 1 . 16 mg/ml , agitated overnight , and filter sterilized using a 0 . 22-µm filter . To replete cholesterol , MDDCs were incubated with cholesterol-saturated methyl-β-cyclodextrin at a concentration of 300 µM cholesterol for 30 minutes at 37°C and then washed five times with PBS before being used in virus capture and transfer studies . MDDCs ( 2 . 5×105 cells ) were seeded above a Transwell insert with a 5 µm pore size and allowed to migrate through the insert in response to medium or CCL21 ( PeproTech ) . Cells above and below the Transwell insert were fixed in 2% paraformaldehyde and counted in a hemocytometer to determine the relative migratory capacity of the MDDCs . Migration index was calculated by dividing the number of experimental cells that migrated in response to CCL21 by the number of untreated cells that migrated in response to media alone . Cholesterol efflux into cell-free culture supernatants and cholesterol content of lysed MDDCs were measured using the AmplexRed cholesterol assay kit per the manufacturer's instructions ( Invitrogen ) . MDDCs were transfected with plasmids that encoded either a mixture of three to five shRNAs directed against ABCA1 or a mixture of control shRNAs ( Santa Cruz Biotechnology ) and a puromycin-resistance gene using Oligofectamine ( Invitrogen ) per the manufacturer's instructions . Transfected cells were selected by culture in the presence of puromycin for 48 hours and then used for cholesterol efflux assays , used for HIV-1 capture assays , or lysed for immunoblot analysis to measure ABCA1 expression . MDDC phenotypes were assessed using antibodies against HLA-DR , CD80 , CD86 , DC-SIGN , CD11c , CD4 , CCR5 , CXCR4 , and CCR7 . Primary CD4+ T cell phenotypes were assessed using antibodies to CD3 , CD4 , CD8 , CD45RO , CD45RA , CCR5 , and CXCR4 . Flow cytometric data was acquired using a Becton-Dickenson FACScan II and data was analyzed using FlowJo software . MDDCs ( 2 . 5×105 cells/well ) or pDCs ( 1×105 cells/well ) were treated with PAM3CSK4 ( 100 ng/ml ) , LPS ( 100 ng/ml ) , CLO97 ( 1 µg/ml ) , or CpG ODN 2006 ( 5 µM ) for 24 hours in the presence or absence of nuclear receptor ligands as described in the legend to figure 1 . Cell-free culture supernatants were collected and analyzed for TNF-α ( eBioscience ) , IL-6 ( eBioscience ) , IL-8 ( BioLegend ) , MIP-1α ( PeproTech ) , and RANTES ( PeproTech ) release by commercially-available ELISA following the manufacturer's instructions . MDDC cell viability was assessed by trypan blue dye exclusion , MTT cytotoxicity assay , and LDH release using a commercial kit ( Promega ) per the manufacturer's instructions . Untreated control and ligand-treated experimental samples were compared using a two-tailed t-test . Experiments were performed in duplicate ( mDCs and pDCs ) or triplicate ( MDDCs ) using cells from a minimum of three different donors as indicated in the figure legends ( n ) . Data are presented as the mean ± standard deviation of pooled data from at least three donors . PPARγ Swiss-Prot # P37231; LXRα Swiss-Prot # Q13133; LXRβ Swiss-Prot # P55055; CCR7 Swiss-Prot # P32248; CCL21 Swiss-Prot # O00585;TLR1 Swiss-Prot # Q15399; TLR2 Swiss-Prot # O60603; TLR4 Swiss-Prot # O00206; TLR7 Swiss-Prot # Q9NYK1; TLR9 Swiss-Prot # C3W5P5; ABCA1 Swiss-Prot # O95477; ABCG1 Swiss-Prot # P45844 .
|
Heterosexual transmission is the primary mode of HIV transmission worldwide . In the absence of an effective vaccine , there is an increasing demand for the development of effective microbicides that block HIV sexual transmission . Dendritic cells ( DCs ) play a critical role in HIV transmission by efficiently binding virus particles , migrating to lymph nodes , and transmitting them to CD4+ T cells , a process called trans-infection . In addition , DCs secrete proinflammatory cytokines that create a favorable environment for virus replication . DC maturation by pathogen-encoded TLR ligands or proinflammatory cytokines dramatically increases their capacity to capture HIV , migrate to lymphoid tissue , and trans-infect T cells . Here , we report that signaling through the nuclear receptors PPARγ and LXR prevents DC maturation and proinflammatory cytokine production , as well as migration . In addition , PPARγ and LXR signaling prevents efficient DC capture and transfer of infectious HIV by increasing ABCA1-mediated cholesterol efflux . Our studies suggest that PPARγ and LXR may be targets for drugs that can inhibit specific aspects of HIV mucosal transmission , namely inflammation , migration , and virus capture and transfer . These findings provide a rationale for considering PPARγ and LXR agonists as potential combination therapies with conventional anti-viral microbicides that target other aspects of mucosal HIV transmission .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"virology/immunodeficiency",
"viruses",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"virology/new",
"therapies,",
"including",
"antivirals",
"and",
"immunotherapy",
"immunology/innate",
"immunity"
] |
2010
|
PPARγ and LXR Signaling Inhibit Dendritic Cell-Mediated HIV-1 Capture and trans-Infection
|
We statistically characterize the population spiking activity obtained from simultaneous recordings of neurons across all layers of a cortical microcolumn . Three types of models are compared: an Ising model which captures pairwise correlations between units , a Restricted Boltzmann Machine ( RBM ) which allows for modeling of higher-order correlations , and a semi-Restricted Boltzmann Machine which is a combination of Ising and RBM models . Model parameters were estimated in a fast and efficient manner using minimum probability flow , and log likelihoods were compared using annealed importance sampling . The higher-order models reveal localized activity patterns which reflect the laminar organization of neurons within a cortical column . The higher-order models also outperformed the Ising model in log-likelihood: On populations of 20 cells , the RBM had 10% higher log-likelihood ( relative to an independent model ) than a pairwise model , increasing to 45% gain in a larger network with 100 spatiotemporal elements , consisting of 10 neurons over 10 time steps . We further removed the need to model stimulus-induced correlations by incorporating a peri-stimulus time histogram term , in which case the higher order models continued to perform best . These results demonstrate the importance of higher-order interactions to describe the structure of correlated activity in cortical networks . Boltzmann Machines with hidden units provide a succinct and effective way to capture these dependencies without increasing the difficulty of model estimation and evaluation .
Electrophysiology is rapidly moving towards high density recording techniques capable of capturing the simultaneous activity of large populations of neurons . This raises the challenge of understanding how networks encode and process information in ways that go beyond tuning properties or feedforward receptive field models . Modeling the distribution of states in a network provides a way to discover communication patterns between neurons or functional groupings such as cell assemblies which may exhibit a more direct relation to stimulus or behavioral variables . The Ising model , originally developed in the 1920s to describe magnetic interactions [1] , has been used to statistically characterize electrophysiological data , particularly in the retina [2] , and more recently for cortical recordings [3] , [4] . This model treats spikes from a population of neurons binned in time as binary vectors and captures dependencies between cells with the maximum entropy distribution for pairwise dependencies . This has been shown to provide a good model for small groups of cells in the retina [5] , though it is unable to capture dependencies higher than second-order . In this work , we apply maximum entropy models to neural population recordings from the visual cortex . Cortical networks have proven more challenging to model than the retina: The magnitude and importance of pairwise correlations between cortical cells is controversial [6] , [7] and higher-order correlations , i . e . correlations which cannot be captured by a pair-wise maximum entropy model , play a more important role [8]–[10] . One of the challenges with current recording technologies is that we can record simultaneously only a tiny fraction of the cells that make up a cortical circuit . Sparse sampling together with the complexity of the circuit mean that the majority of a cell's input will be from cells outside the recorded population . In adult cat visual cortex , direct synaptic connections have been reported to occur between 11%–30% of nearby pairs of excitatory neurons in layer IV [11] , while a larger fraction of cell pairs show “polysynaptic” couplings [12] , defined by a broad peak in the cross-correlation between two cells . This type of coupling can be due to common inputs ( either from a different cortical area or lateral connections ) or a chain of monosynaptic connections . A combination of these is believed to give rise to most of the statistical interactions between recorded pairs of cells . The Ising model , which assumes only pairwise couplings , is well suited to model direct ( and symmetric ) synaptic coupling , but cannot capture interactions involving more than two cells . We propose a new approach , that addresses both incomplete sampling and common inputs from other cell assemblies , by extending the Ising model with a layer of hidden units or latent variables . The resulting model is a semi-Restricted Boltzmann Machine ( sRBM ) , which combines pairwise connections between visible units with an additional set of connections to hidden units . Estimating the parameters of energy-based models , to which Ising models and Boltzmann machines belong , is computationally hard because these models cannot be normalized in closed form . For both Ising models and Boltzmann machines with hidden units , the normalization constant is intractable to compute , consisting of a sum over the exponential number of states of the system . This makes exact maximum likelihood estimation impossible for all but the smallest systems and necessitates approximate or computationally expensive estimation methods . In this work , we use Minimum Probability Flow ( MPF [13] , [14] , in the context of neural decoding see [4] , [15] ) to estimate parameters efficiently without computing the intractable partition function . It provides a straightforward way to estimate the parameters of Ising models and Boltzmann machines for high-dimensional data . Another challenge in using energy-based models is the evaluation of their likelihood after fitting to the data , which is again made difficult due to the partition function . To compute probabilities and compare the likelihood of different models , annealed importance sampling ( AIS ) [16] was used to estimate the partition function . Combining these two methods for model estimation and evaluation , we show that with hidden units , Boltzmann machines can capture the distribution of states in a microcolumn of cat visual cortex significantly better than an Ising model without hidden units . The higher-order structure discovered by the model is spatially organized and specific to cortical layers , indicating that common input or recurrent connectivity within individual layers of a microcolumn are the dominant source of correlations . Applied to spatiotemporal patterns of activity , the model captures temporal structure in addition to dependencies across different cells , allowing us to predict spiking activity based on the history of the network .
We estimated Ising , RBM and sRBM models for populations of cortical cells simultaneously recorded across all cortical layers in a microcolumn of cat V1 in response to long , continuous natural movies presented at a frame rate of 150 Hz . Code for the model estimation is available for download at http://github . com/ursk/srbm . Fig . 1a ) shows an example frame from one of the movies . Model parameters were estimated using MPF with an regularization penalty on the model parameters to prevent overfitting . To compute and compare likelihoods , the models were normalized using AIS . Here we present data from two animals , one with 22 single units ( B4 ) , another with 36 units ( T6 ) , as well as a multiunit recording with 26 units ( B4M ) . Fig . 1b ) shows spiking data from session B4 in 20 ms bins , with black squares indicating a spike in a bin . Spatiotemporal datasets were constructed by concatenating spikes from consecutive time bins . Pairs of cells show weak positive correlations , shown in Fig . 1c ) , and noise correlations computed from 60 repetitions of a 30s stimulus are similarly small and positive . For all recordings , the population was verified to be visually responsive and the majority of cells were orientation selective simple or complex cells . As recordings were performed from a single cortical column , receptive fields shared the same retinotopic location and have similar orientation selectivity , differing mostly in size , spatial frequency and phase selectivity . See [17] for a receptive field analysis performed on the same raw data . The estimated model parameters for the three different types of models ( Ising , RBM and sRBM ) are shown in Fig . 2 for session B4 . The sparseness penalty , chosen to optimize likelihood on a validation dataset , results in many of the parameters being zero . For the Ising model ( a ) we show the coupling as a matrix plot , with lines indicating anatomical layer boundaries . The diagonal contains the bias terms , which are negative since all cells are off the majority of the time . The matrix has many small positive weights that encourage positive pairwise correlations . In ( b ) we show the hidden units of the RBM as individual bar plots , with the bars representing connection strengths to visible units . The topmost bar corresponds to the hidden bias of the unit , and hidden units are ordered from highest to lowest variance . The units are highly selective in connectivity: The first unit almost exclusively connects to cells in the deep ( granular and subgranular ) cortical layers . The second unit captures correlations between cells in the superficial ( supergranular ) layers . The correlations are of high order , with 10 and more cells receiving input from a hidden unit . The remaining units connect fewer cells , but still tend to be location-specific . Only the hidden units that have non-zero couplings are shown . Additional hidden units are turned off by the sparseness penalty , which was chosen to maximize likelihood on the cross-validation dataset . The interpretation of hidden units is quite similar to the pairwise terms of the Ising model: positive coupling to a group of visible units encourages these units to become active simultaneously , as the energy of the system is lowered if both the hidden unit and the cells it connects to are active . Thus the hidden units become switched on when cells they connect to are firing ( activation of hidden units not shown ) . The sRBM combines both pairwise and hidden connections and hence is visualized with a pairwise coupling matrix and bar plots for hidden units . With the larger number of parameters , the best model is even more sparse in the number of nonzero parameters . The remaining pairwise terms predominantly encode negative interactions , and much of the positive coupling has been explained away by the hidden units . These give rise to strong positive couplings within either superficial ( II/III ) or intermediate ( IV ) and deep ( V/VI ) layers , which explain the majority of structure in the data . The more succinct explanation for dependencies between recorded neurons is via connections to shared hidden units , rather than direct couplings between visible units . The RBM and sRBM in this comparison were both estimated with 22 hidden units , but we show only units that did not turn off entirely due to the sparseness penalty . In this example , a sparseness penalty of was found to be optimal for all three models . In order to ascertain to what degree the stimulus driven component of activity accounts for the learned higher-order correlations , we augmented the above models with a dynamic bias term that consists of the log of the average instantaneous firing probability of each cell over repeated presentations of the same stimulus . In the case that all trained parameters were zero , this model would assign a firing probability to all neurons identical to that in the peri-stimulus time histogram ( PSTH ) . In Fig . 2d ) the couplings for the Ising model with stimulus terms are shown . As the pairwise couplings now only capture additional structure beyond correlations explained by the stimulus , they tend to be weaker than in the Ising model without stimulus terms . In particular the bias terms on the diagonal are almost completely explained away by the dynamic bias . The same reasoning applies to the RBM with PSTH terms , which is shown in e ) . Although the couplings are weaker than for the pure RBM , the basic structure remains , with the first two hidden units explaining correlations within superficial and deep groups of cells , respectively . This shows that the learned coupling structure can not be explained purely from higher-order stimulus correlations and receptive field overlap . Even when stimulus-induced correlations are fully accounted for , the correlation structure captured by the RBM remains similar and higher-order correlations are the dominant driver of correlated firing . For a quantitative comparison between models , we computed normalized likelihoods using Annealed Importance Sampling ( AIS ) to estimate the partition function . For each model , we generated 500 samples through a chain of annealing steps . To ensure convergence of the chain , we use a series of chains varying the number of annealing steps and verify that the estimate of the partition function stabilizes to within at least 0 . 02 bits/s ( see Fig . S1 ) . For models of size 20 and smaller we furthermore computed the partition function exactly to verify the AIS estimate . Fig . 3a ) shows a comparison of excess log likelihood for the three different models and on all three datasets . , which we define as the gain in likelihood over the independent firing rate model , is computed in units of bits/spike for the full population . Both higher-order models outperform the Ising model in fitting the datasets significantly . Error bars are standard deviation computed from 10 models with different random subsets of the data used for learning and validation , and different random seeds for the parameters and the AIS sampling runs . Fig . 3b ) shows the excess log likelihood for the models with stimulus terms . Due to the additional computational complexity , these models were only estimated for the small B4 data set . The left of the two bar plots shows that including the stimulus information through the PSTH greatly increases the likelihood , even the PSTH only model without coupling terms outperforms the Ising and RBM models by about 0 . 7 bits/s . Including coupling terms still increases the likelihood , which is particularly visible on the right bar plot which shows the log likelihood gain relative to the PSTH model . Including higher-order coupling terms still provides a significant gain over the pairwise model , confirming that there are higher-order correlation in the data beyond those induced by the stimulus . Each of the models was estimated for a range of sparseness parameters bracketing the optimal using 4-fold cross-validation on a holdout set , and the results are shown for the optimal choice of for each model . Additional insight into the relative performance of the models can be gained by comparing model probabilities to empirical probabilities for the various types of patterns . Fig . 4 shows scatter plots of model probabilities under the different models against pattern frequencies in the data . Patterns with a single active cell , two simultaneously active cells , etc . are distinguished by different symbols . As expected from the positive correlations , the independent model ( yellow ) shown in a ) consistently overestimates the probabilities of cells being active individually , so these patterns fall above the identity line , while all other patterns are underestimated . For comparison , the Ising model is shown in the same plot ( blue ) , and does significantly better , indicated by the points moving closer to the identity line . It still tends to fail in a similar way though , with many of the triplet patterns being underestimated as the model cannot capture triplet correlations . In b ) , this model is directly compared against the RBM ( green ) . Except for very rare patterns , most points are now very close to the identity line , as the model can fully capture higher-order dependencies . Hidden units describe global dependencies that greatly increase the frequency of high order patterns compared to individually active cells . The 5% and 95% confidence intervals for the counting noise expected in the empirical frequency of states are shown as dashed lines . The solid line is the identity . Inserts in both models show the distribution of synchrony , , where is the number of cells simultaneously firing in one time bin . This metric has been used for example in [18] to show how pairwise models fail to capture higher-order dependencies . In the case of the T6 data set with 36 cells shown here , the Ising model and RBM both provide a good fit to the distribution of synchrony in the observed data . Note that any error in estimating the partition function of the models would lead to a vertical offset of all points . Thus visually checking the alignment of the data cloud around the identity line provides a visual verification that there are no catastrophic errors in the estimation of the partition function . Unfortunately we cannot use this alignment as a shortcut to compute the partition function without sampling , e . g . by defining such that the all zeros state has the correct frequency , as this assumes a perfect model fit . For instance , regularization tends to reduce model probabilities of the most frequent states , so this estimate of would systematically overestimate the likelihood of regularized models . We note , however , that for higher-order models with no regularization this estimate does indeed agree well with the AIS estimate . The same models can be used to capture spatiotemporal patterns by treating previous time steps as additional cells . Consecutive network states binned at 6 . 7 ms were concatenated in blocks of up to 13 time steps , for a total network dimensionality of 130 with 10 cells . These models were cross-validated and the sparseness parameters optimized in the same way as for the instantaneous model . This allows us to learn kernels that describe the temporal structure of interactions between cells . In Fig . 5 we compare the relative performance of spatiotemporal Ising and higher-order models as a function of the number of time steps included in the model . To create the datasets , we picked a subset of 10 cells with the highest firing rates from the B4 dataset ( 4 cells from subgranular , 2 from granuar and 4 from supergranular layers ) and concatenated blocks of up to 13 subsequent data vectors . This way models of any dimensionality divisible by 10 can be estimated . The number of parameters of the RBM and Ising model were kept the same by fixing the number of hidden units in the RBM to be equal to the number of visible units , the sRBM was also estimated with a square weight matrix for the hidden layer . As before , the higher-order models consistently outperform the Ising model . The likelihood per spike increases with the network size for all models , as additional information from network interactions leads to an improvement in the predictive power of the model . The curve for the Ising model levels off after a dimensionality of about 30 is reached , as higher-order structure that is not well captured by pairwise coupling becomes increasingly important . However , the likelihood of higher-order models continues to increase through the entire experimental range . The insert in the figure shows the entropy of the models , normalized by the data dimensionality by dividing by the number of frames and neurons . The entropy was computed as where the expectation of the energy was estimated by initializing 100 , 000 samples using the holdout data set , and then running 2000 steps of Gibbs sampling . Due to temporal dependencies additional frames carry less entropy , but we do not reach the point of extensivity where the additional entropy per frame reaches a constant value . As the RBM is better able to explain additional structure in new frames , the additional entropy for new frames is much less than for the Ising model . A similar observation has been made in [5] , where Ising and higher-order models for 100 retinal ganglion cells were compared to models for 10 time steps of 10 cells . It is noteworthy that temporal dependencies are similar to dependencies between different cells , in that there are strong higher-order correlations not well described by pairwise couplings . These dependencies extend surprisingly far across time ( at least 87 ms , corresponding to the largest models estimated here ) and are of such a form that including pairwise couplings to these states does not increase the likelihood of the model . This has implications e . g . for GLMs that are typically estimated with linear spike coupling kernels which will likely miss these interactions . To predict spiking based on the network history , we can compute the conditional distribution of single units given the state of the rest of the network . This is illustrated for a network with 15 time steps for a dimensionality of 150 . This model is not included in the above likelihood comparison , as the AIS normalization becomes very expensive for this model size . Fig . 6a ) shows the learned weights of 18 randomly chosen nonzero hidden units for a spatiotemporal RBM model with 150 hidden units . Each subplot corresponds to one hidden unit , which connects to 10 neurons ( vertical axis ) across 15 time steps or 100 ms ( horizontal axis ) . Some units specialize in spatial coupling across different cells at a constant time lag . As the model has no explicit notion of time , the time lag of these spatial couplings is not unique and the model learns multiple copies of the same spatial pattern . Thus while there are 55 nonzero hidden units , the number of unique patterns is much smaller so that the effective representation is quite sparse . The remaining units describe smooth , long-range temporal dependencies , typically for small groups of cells . Both of these subpopulations capture higher-order structure connecting many neurons that cannot be well approximated with pairwise couplings . By conditioning the probability of one cell at one time bin on the state of the remaining network , we can compute how much information about a cell is captured by the model over a naive prediction based on the firing rate of the cell . This conditional likelihood for each cell is plotted in Fig . 6b ) in a similar way to excess log likelihood for the entire population in Fig . 5 , except in units of bits per second rather than bits per spike . While the result here reflects our previous observation that Boltzmann machines with hidden units outperform Ising models , we note that the conditional probabilities are easily normalized in closed form since they describe a one-dimensional state space . Thus we can ensure that the likelihood gain holds independent of the estimation of and is not due to systematic errors in sampling from the high-dimensional models . Fig . 6c ) provides a more intuitive look at the prediction . For 1 s of data from one cell , where 5 spikes occur , we show the conditional firing probabilities for the three models given 100 ms of history of itself and the other cells . Qualitatively , the models perform well in predicting spiking probabilities , suggesting it might compare favorably to prediction based on GLMs or Ising models [19] .
While there has been a resurgence of interest in Ising-like maximum entropy models for describing neural data , progress has been hampered mainly by two problems . First , estimation of energy based models is difficult since these models cannot be normalized in closed form . Evaluating the likelihood of the model thus requires approximations or a numerical integral over the exponential number of states of the model , making maximum likelihood estimation computationally intractable . Even the pairwise Ising model is typically intractable to estimate , and various approximations are required to overcome this problem . Second , the number of model parameters to be estimated grows very rapidly with neural population size . If correlations up to order are considered , the number of parameters is proportional to . In general , fully describing the distribution over states requires a number of parameters which is exponential in the number of neurons . This can be dealt with by cutting off dependencies at some low order , by estimating only a small number of higher-order coupling terms , or by imposing some specific form on the dependencies . We attempted to address both of these problems here . Parameter estimation was made tractable using MPF , and latent variables were shown to be an effective way of capturing high order dependencies . This addresses several shortcomings that have been identified with the Ising model . As argued in [20] , models with direct ( pairwise ) couplings are not well suited to model data recorded from cortical networks . Since only a tiny fraction of the neurons making up the circuit are recorded , most input is likely to be common input to many of the recorded cells rather than direct synapses between them . While this work compares Generalized Linear Models ( GLMs ) such as models of the retina [21] and for LGN [22] to linear dynamical systems ( LDS ) models , the argument applies equally for the models presented here . Another shortcoming of the Ising model and some previous extensions is that the number of parameters to be estimated does not scale favorably with the dimensionality of the network . The number of pairwise coupling terms in GLM and Ising models scales with the square of the number of neurons , so with the amounts of data typically collected in electrophysiological experiments it is only possible to identify the parameters for small networks with a few tens of cells . This problem is aggravated by including higher-order couplings: for example the number of third order coupling parameters scales with the cube of the data dimensionality . Therefore attempting to estimate these coupling parameters directly is a daunting task that usually requires approximations and strong regularization . Early attempts at modeling higher-order structure side-stepped these technical issues by focussing on structure in very small networks . Ohiorhenuan noted that Ising models fail to explain structure in cat visual cortex [8] and was able to model triplet correlations [9] by considering very small populations of no more than 6 neurons . Similarly , Yu et al . [3] , [10] show that over the scale of adjacent cortical columns of anesthetized cat visual cortex , small subnetworks of 10 cells are better characterized with a dichotomized Gaussian model than the pairwise maximum entropy distribution . While the dichotomized Gaussian [23] is estimated only from pairwise statistics , it carries higher-order correlations that can be interpreted as common Gaussian inputs [24] . However these correlations are implicit in the structure of the model and not directly estimated from the data as with the RBM , so it is not clear that the model would perform as well on different datasets . Given that higher-order correlations are important to include in statistical models of neural activity , the question turns to how these models can be estimated for larger data sets . In this section , we focus on two approaches that are complementary to our model using hidden units . The increasing role of higher-order correlations in larger networks was first observed in [25] , where Ising models were fit via MCMC methods to the same 40 cell retina dataset that was analyzed in terms of subsets of 10 cells in [2] . This point is further emphasized by Schneidman and Ganmor in [5] , who caution that trying to model small subsets ( 10 cells ) of a larger network to infer properties of the full network may lead to incorrect conclusions , and show that for retinal networks higher-order correlations start to dominate only once a certain network size is reached . Therefore they address the same question as the present paper , i . e . how to capture order correlations without the accompanying growth in the number of free parameters in a larger network . In their proposed Reliable Interaction Model ( RIM ) , they exploit the sparseness of the neural firing patterns to argue that it is possible to explicitly include third , forth and higher-order terms in the distribution , as most higher-order coupling terms will be zero . Therefore the true distribution can be well approximated from a small number of these terms , which can be calculated using a simple recursive scheme . In practice , the main caveat is that only patterns that appear in the data many times are used to calculate the coupling terms . While the model by construction assigns correct relative probabilities to observed patterns , the probability assigned to unobserved patterns is unconstrained , and most of the RIM's probability mass may thus be assigned to states which never occur in the data . The second alternative to the RBM with hidden units is to include additional low-dimensional constraints in an Ising model . In the “K-pairwise” model [18] , [26] , in addition to constraining pairwise correlations , a term is introduced to constrain the probability of neurons being active simultaneously . This adds very little model complexity , but significantly improves the model of the data . This is shown , for example , by computing conditional predictions in a similar fashion to that shown in Fig . 6c ) , where the K-pairwise model for a population of 100 retinal ganglion cells has an 80% correlation with the repeated trial PSTH . In contrast to the RBM , however , this model is not structured in a way that can be easily interpreted in terms of functional connectivity . To estimate these models for large numbers ( ) of neurons , the authors leverage the sampling algorithm described in [27] , an -regularized histogram Monte Carlo method . In addition to proposing a faster ( though slower than MPF ) parameter estimation method for this class of models , Tkačik and colleagues address the difficulty in sampling from the model and computing the partition function . In our experiments the overall limiting factor is the Gibbs sampler in the AIS partition function estimation . Tkačik et al . use a more efficient sampling algorithm ( Wang-Landau ) to compute partition functions and entropy of their models . As an even simpler approach to the partition function problem , they suggest that it can be obtained in closed form if the empirical probability of at least one pattern in the data is accurately known . A case in point is the all zeros pattern that is typically frequent for recordings with sparsely firing neurons . Unfortunately , this approach is limited in that it assumes that the probability the model assigns to the state is identical to the empirical probability of the state . In the case that the model has not been perfectly fit to the data , or in the case that the data does not belong to the model class , this will lead to an incorrect estimate of the partition function . Since the activity we are modeling is in response to a specific stimulus , one may rightfully question whether the observed higher-order correlations in neural activity are simply due to higher-order structure contained in the stimulus , as opposed to being an emergent property of cortical networks . In an attempt to tease apart the contribution of the stimulus , we included a nonparametric PSTH term in the model . However , this can capture arbitrarily complex stimulus transformations using the trial-averaged response to predict the response to a new repetition of the same stimulus . As an “oracle model” , it does not only capture the part of the response that could be attributed to a feed-forward receptive field , but also captures contextual modulation effects mediated by surrounding columns and feedback from higher brain areas , essentially making it “too good” as a stimulus model . The RBM and Ising models are then relegated to merely explain the trial to trial variability in our experiments . Not including stimulus terms and finding the best model to explain the correlations present in the data , irrespective of whether they are due to stimulus or correlated variability , seems to be an equally valid approach to discover functional connectivity in the population . GLMs [21] can be used to model each cell conditioned on the rest of the population . While mostly used for stimulus response models including stimulus terms , they are easily extended with terms for cross-spike coupling , which capture interactions between cells . GLMs have been successfully augmented with latent variables [20] , for instance to model the effect of common noisy inputs on synchronized firing at fast timescales [28] . A major limitation of GLMs is that current implementations can only be estimated efficiently if they are linear in the stimulus features and network coupling terms , so they are not easily generalized to higher-order interactions . Two approaches have been used to overcome this limitation for stimulus terms . The GLM can be extended with additional nonlinearities , preserving convexity on subproblems [22] . Alternatively , the stimulus terms can be packaged into nonlinear features which are computed in preprocessing and usually come with the penalty of a large increase in the dimensionality of the problem [29] . However , we are not aware of any work applying either of these ideas to spike history rather than stimulus terms . Another noteworthy drawback of GLMs is that instantaneous coupling terms cannot be included [20] , so instantaneous correlations cannot be modeled and have to be approximated using very fine temporal discretization . The RBM provides a parsimonious model for higher-order dependencies in neural population data . Without explicitly enumerating a potentially exponential number of coupling terms or being constrained by only measurements of pairwise correlations , it provides a low-dimensional , physiologically interpretable model that can be easily estimated for populations of 100 or more cells . The connectivity patterns the RBM learns from cells simultaneously recorded from all cortical layers are spatially localized , showing that small neural assemblies within cortical layers are strongly coupled . This suggests that cells within a layer perform similar computations on common input , while cells across different cortical layers participate in distinct computations and have much less coupled activity . This novel observation is made possible by the RBM: because each of the hidden units responds to ( and therefore learns on ) a large number of recorded patterns , it can capture dependencies that are too weak to extract with previous models . In particular , the connectivity patterns discovered by the RBM and sRBM are by no means obvious from the covariance of the data or by inspecting the coupling matrix of the Ising model . This approach , combining a straightforward estimation procedure and a powerful model , can be extended from polytrode recordings to capture physiologically meaningful connectivity patterns in other types of multi-electrode data .
The sRBM consists of a set of binary visible units corresponding to observed neurons in the data and a set of hidden units that capture higher-order dependencies . Weights between visible units , corresponding to an Ising model or fully visible Boltzmann machine , capture pairwise couplings in the data . Weights between visible and hidden units , corresponding to an RBM , learn to describe higher-order structure . The Ising model with visible-visible coupling weights and biases has an energy function ( 1 ) with associated probability distribution , where the normalization constant , or partition function , consists of a sum over all system states . The RBM with visible-hidden coupling weights and hidden and visible biases and has an energy function ( 2 ) with associated probability distribution . Since the there are no connections between hidden units ( hence “restricted” Boltzmann machine ) , the latent variables can be analytically marginalized out of the distribution ( see supplementary information ) to obtain ( 3 ) This step gives a standard energy-based model which we can estimate in our framework , while in a fully connected Boltzmann machine we could not marginalize over hidden units , making the estimation intractable . The energy for the marginalized distribution over ( sometimes referred to as the free energy in machine learning literature ) is ( 4 ) where is the row of the matrix . The energy function for an sRBM combines the Ising model and RBM energy terms , ( 5 ) As with the RBM , it is straightforward to marginalize over the hidden units for an sRBM , ( 6 ) ( 7 ) A hierarchical Markov Random Field based on the sRBM has previously been applied as a model for natural image patches [34] , with the parameters estimated using contrastive divergence ( CD ) [35] . To include stimulus effects into the Boltzmann machine models , we start with a maximum entropy model constrained to fit the peri-stimulus time histogram ( PSTH ) , i . e . the response to a given stimulus obtained by empirically computing the firing probabilities averaged over repeated presentations . This non-parametric model has the form ( 8 ) where the subscript refers to histogram , and both the dynamic bias term and the data vector have explicit time dependence . The dynamic bias terms in this model can be computed in closed form where the PSTH sums over stimulus repetitions and counts the number of spikes fired by the neuron . Starting with this model and keeping the dynamic bias term fixed at the closed form solution , we add pairwise and higher-order coupling terms to obtain ( 9 ) ( 10 ) which combine the Ising and RBM population models , respectively , with the stimulus model given by the PSTH term . Estimation of these models is no more difficult than the standard Ising and RBM models , since the PSTH term is computed in closed form and effectively only adds a constant to the energy function . This method corresponds to model T2 ( explicit time dependence , second order stimulus dependence ) of [36] , where it is not further explored due to the requirement for repeated stimulus segments . We used 60 repetitions of a 30 s long natural movie to estimate the PSTH . Instead of CD , which is based on sampling , we train the models using Minimum Probability Flow ( MPF , [13] ) , a recently developed parameter estimation method for energy based models . MPF works by minimizing the KL divergence between the data distribution and the distribution which results from moving slightly away from the data distribution towards the model distribution . This KL divergence will be uniquely zero in the case where the model distribution is identical to the data distribution . While CD is a stochastic heuristic for parameter update , MPF provides a deterministic and easy to evaluate objective function . Second order gradient methods can therefore be used to speed up optimization considerably . The MPF objective function ( 11 ) measures the flow of probability out of data states into neighboring non-data states , where the connectivity function identifies neighboring states , and is the list of data states . We consider the case where the connectivity function is set to connect all states which differ by a single bit flip ( 12 ) where is the Hamming distance between and . See supplementary information for a derivation of the MPF objective function and gradients for the sRBM , RBM , and Ising models . In all experiments , minimization of K was performed with the MinFunc implementation of L-BFGS [37] . To prevent overfitting all models were estimated with an sparseness penalty on the coupling parameters . This was done by adding a term of the form to the objective function , summing over the absolute values of the elements of both the visible and hidden weight matrices . The optimal sparseness was chosen by cross-validating the log-likelihood on a holdout set , with the value of selected from the set spanning the range of optimal regularization for all models . Since MPF learning does not give an estimate of the partition function , for models that were too large to normalize by summing over all states , we use annealed importance sampling ( AIS , [16] ) to compute normalized probabilities . AIS is a sequential Monte Carlo method that works by gradually morphing a distribution with a known normalization constant ( in our case a uniform distribution over ) into the distribution of interest . See supplementary information for more detail . AIS applied to RBM models is described in [38] , which also highlights the shortcomings of previously used deterministic approximations . Since AIS requires running a sampling chain , in our case a Gibbs sampler , it generally takes much longer than the parameter estimation . Normalizing the distribution via AIS allows us to compute the log likelihood of the model and compare it to the likelihood gained over a baseline model . This baseline assumes cells to be independent and characterized by their firing rate with rates for individual cells . The independent model is easily estimated and normalized , and is commonly used as a reference for model comparison . The excess log likelihood over this baseline is defined in terms of a sample expectation as . The excess log likelihood rate , computed in bits/spike by normalizing with the population firing rate per time bin , is used as the basis for model comparisons . Normalizing by the firing rate and comparing bits/spike rather than bits per unit time has the advantage that this measure is less sensitive to the overall activity when comparing across data sets . For models with stimulus terms , a separate normalization constant is required for each state of the dynamic bias . As our 30 s stimulus segment data set consists of 1500 time bins , this amounts to computing 1500 separate partition functions . This limits us to small models with up to 20 neurons where the partition function can be computed quickly by enumerating the full state space , avoiding the use of the costly AIS sampling procedure . To compare likelihoods , we use the PSTH model ( i . e . the model with only stimulus and no coupling terms ) as the baseline since all three likelihoods are much higher than for models without stimulus terms .
|
Communication between neurons underlies all perception and cognition . Hence , to understand how the brain's sensory systems such as the visual cortex work , we need to model how neurons encode and communicate information about the world . To this end , we simultaneously recorded the activity of many neurons in a cortical column , a fundamental building block of information processing in the brain . This allows us to discover statistical structure in their activity , a first step to uncovering communication pathways and coding principles . To capture the statistical structure of firing patterns , we fit models that assign a probability to each observed pattern . Fitting probability distributions is generally difficult because the model probabilities of all possible states have to sum to one , and enumerating all possible states in a large system is not possible . Making use of recent advances in parameter estimation , we are able to fit models and test the quality of the fit to the data . The resulting model parameters can be interpreted as the effective connectivity between groups of cells , thus revealing patterns of interaction between neurons in a cortical circuit .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"computational",
"neuroscience",
"neuroscience",
"biology",
"and",
"life",
"sciences",
"computational",
"biology"
] |
2014
|
Modeling Higher-Order Correlations within Cortical Microcolumns
|
Seckel syndrome is a recessively inherited dwarfism disorder characterized by microcephaly and a unique head profile . Genetically , it constitutes a heterogeneous condition , with several loci mapped ( SCKL1-5 ) but only three disease genes identified: the ATR , CENPJ , and CEP152 genes that control cellular responses to DNA damage . We previously mapped a Seckel syndrome locus to chromosome 18p11 . 31-q11 . 2 ( SCKL2 ) . Here , we report two mutations in the CtIP ( RBBP8 ) gene within this locus that result in expression of C-terminally truncated forms of CtIP . We propose that these mutations are the molecular cause of the disease observed in the previously described SCKL2 family and in an additional unrelated family diagnosed with a similar form of congenital microcephaly termed Jawad syndrome . While an exonic frameshift mutation was found in the Jawad family , the SCKL2 family carries a splicing mutation that yields a dominant-negative form of CtIP . Further characterization of cell lines derived from the SCKL2 family revealed defective DNA damage induced formation of single-stranded DNA , a critical co-factor for ATR activation . Accordingly , SCKL2 cells present a lowered apoptopic threshold and hypersensitivity to DNA damage . Notably , over-expression of a comparable truncated CtIP variant in non-Seckel cells recapitulates SCKL2 cellular phenotypes in a dose-dependent manner . This work thus identifies CtIP as a disease gene for Seckel and Jawad syndromes and defines a new type of genetic disease mechanism in which a dominant negative mutation yields a recessively inherited disorder .
Seckel syndrome ( SS ) belongs to the group of genome instability disorders collectively referred to as DNA-damage response ( DDR ) and repair defective syndromes [1] . While cancer predisposition is often associated with such syndromes , only a few cancers have been reported for SS patients . Instead , SS pathogenesis is primarily based on marked growth and neurological impairments . Moreover , in contrast to some other repair defective syndromes , SS is a heterogeneous disease with five independent loci identified: SCKL1 , which bears a mutation that creates an alternative splicing site in the ATR gene [2]; SCKL2 , previously mapped by us in the chromosomal region 18p11 . 31-q11 . 2 [3]; SCKL3 , mapped in the region 14q23-q24 [4]; SCKL4 that has a mutation in the CENPJ gene [5]; and the recently reported SCKL5 that harbors mutations in CEP152 [6] . Cells derived from all Seckel patients are impaired in signaling mediated by the DNA-damage responsive protein kinase ATR . Thus , SS cells display reduced phosphorylation of downstream ATR substrates , which include the Chk1 effector checkpoint kinase , and impaired G2/M cell-cycle checkpoint arrest upon treatment with UV light or replication blocking agents [7] , [8] . Except in SCKL1 patients where ATR itself is mutated , the connections between the other SCKL loci and ATR activation are not yet clear . ATR is recruited to and activated by replication protein A ( RPA ) -coated single-stranded DNA ( ssDNA ) [7] , which arises by uncoupling of DNA polymerases and helicases at stalled DNA replication forks [9] , [10] or upon processing of DNA double-strand breaks ( DSBs ) [7] , [9] , [11] . DSB processing occurs by DNA-end resection , a 5′-3′ degradation of one of the DNA strands . This process is a tightly regulated and serves as a molecular switch between signaling mediated by the ATM and ATR kinases , and also regulates the way DSBs are repaired [9] . Specifically , resection only takes place effectively in S and G2 phases of the cell cycle , and it is initiated by the combined actions of the MRE11-RAD50-NBS1 ( MRN ) complex and CtIP [11] . The licensing of DNA-end resection requires cell-cycle dependent phosphorylation of CtIP [12] , [13] , and in the absence of CtIP , DSB processing is impaired and ATR activation is hampered [11] . As the above factors suggested that CtIP defects might yield SS , we examined DNA samples from two unrelated microcephalic families that both map to the SCKL2 locus: the original SCKL2 family [3] and a family diagnosed with a Seckel-like type of congenital microcephaly termed Jawad syndrome [14] ( see Figure S1A and S1B ) . As described herein , this analysis revealed that the affected individuals in these families indeed harbor homozygous mutations in the CtIP gene . Strikingly , both mutations lead to premature stop codons in the CtIP transcript and , therefore , to the expression of predicted C-terminal truncation derivatives of CtIP . We show that , while the Jawad two basepair deletion mutation leads to a classical shift in reading frame , the SCKL2 mutation creates an alternative splicing site leading to both the normal and aberrant CtIP proteins coexisting in the cells of patients and carriers . By characterizing SCKL2 cells and CtIP proficient cells artificially expressing a C-terminally truncated CtIP protein , we conclude that , despite being a recessively inherited syndrome , the CtIPSCKL2 mutation encodes a dominant negative protein that impairs ATR activation .
Like other SS cells , SCKL2 cells display defects in ATR signaling in response to DSBs . However , in contrast to all other SS cell lines tested , SCKL2 cells do not exhibit hypersensitivity to replication fork stalling caused by hydroxyurea treatment[8] . As the ATR pathway is activated by ssDNA exposed during polymerase-helicase uncoupling under these circumstances , this implies that ATR and the ATR signaling pathway are functional in SCKL2 cells , and that the molecular defect of SCKL2 cells responding to DSBs is likely upstream of ATR . Moreover , these data suggest that SCKL2 cells might be specifically defective in processing DSBs to ssDNA . Based on these and other criteria , we sequenced the CtIP gene located within the SCKL2 locus ( 18p11 . 31-q11 . 2; Figure 1A ) [3] and found one mutation: a T to G transition 53 bp within the 15th intron ( CtIPs; Figure 1A ) . The mutation co-segregated with the disease in the SCKL2 family and was not present in 100 unrelated control individuals . All seven members of the family were sequenced . No other mutations were found despite in-depth sequencing of the promoter- and untranslated regions ( approximately 3500 bp ) , all coding exons and adjacent intronic sequences ( on average 750 bp ) . Bioinformatics pointed to a 5′ splice-attracting capability for the altered sequence introduced by the CtIPs mutation , suggesting the presence of a competing alternative donor-site within the 15th intron ( Figure S2 ) . An alternative spliced transcript , resulting in an extended exon 15 was thus expected . To assess this possibility , we carried out RT-PCR amplifications targeting both this hypothesized extended region ( Figure 1C , 1D ) and the wild-type transcript sequence on total RNA extracted from EBV-transformed lymphoblasts derived from SCKL2 patients , unaffected family members and unrelated control individuals . No measurable differences in expression were observed between the samples for the wild-type isoform ( not shown ) , but one weakly expressed CtIP transcript , not previously identified , was specifically found in samples from SCKL2 patients ( homozygous ) and at approximately 50% of this level in non-affected family members ( heterozygous; Figure 1B ) . We therefore denoted this transcript CtIPSeckel . We observe uniform expression level of wild-type transcript between cell lines , discarding a reduction in full length CtIP mRNA as a cause for the observed phenotypes in SCKL2 patients and cell lines . Due to the appearance of a premature stop codon , we expected CtIPSeckel transcript to be subjected to NMD and only negligible levels to be expressed . Relative quantification of this mutant transcript against wild type transcript showed in fact a weak expression-ratio of approximately 1∶600 in homozygous SCKL2 cells ( not shown ) . Sequencing the exon-intron junction of CtIPSeckel furthermore confirmed that the naturally occurring CtIP exon 15 donor-site is indeed skipped in this transcript and that RNA-splicing occurs between the introduced donor-site and the exon 16 acceptor-site as predicted ( Figure 1C ) . To test our hypothesis that C-terminal truncation of CtIP can lead to congenital microcephaly , we sequenced DNA samples from several other patients suffering this type of condition . Some of them came from a family diagnosed as suffering from Jawad syndrome ( Figure S1B ) [14] , previously mapped to chromosomal region ( 18p11 . 22-q11 . 2 ) overlapping with the SCKL2 locus [14] . Jawad patients differ from Seckel patients as no growth-impairment has been described for this syndrome . They do , however , share a large number of other characteristics ( see Table 1 for comparison of Seckel and Jawad symptoms ) . As it has been shown that SS and primary microcephaly can be caused by the same genes [15] , [16] , we considered them good candidates for our study . Indeed , in family members with Jawad syndrome , we found a homozygous two base-pair deletion in exon 11 ( CtIPj; Figure 1A ) . This 2 bp deletion causes a frame-shift in the CtIP reading frame and the appearance of a premature stop codon that would yield a truncated CtIP protein ( Figure 1D and Figure S3 ) . We also analyzed DNA from two obligate carriers of the Jawad family and found that both were heterozygous for the mutation . The mutation was not present in any control individual . It is notable that both the CtIPs and CtIPj mutations are predicted to yield C-terminally truncated CtIP proteins ( CtIPSCKL2 and CtIPJawad; Figure 1A and 1D , and Figure S3 ) because C-terminal truncation of CtIP has been shown to cause cells to display defective processing of DNA DSBs and exhibit impaired ATR-dependent signaling [11] . Surprisingly , while CtIPj/j homozygotes presumably only express the truncated form of the protein , CtIPs/s homozygous cells express transcripts for both the truncated and full-length CtIP proteins . We decided to further characterize the mechanisms causing the syndromes at the molecular and cellular levels but , as no cellular or RNA samples were available from the Jawad family , we focused our investigations on the SCKL2 family . According to in silico analyses of the CtIPSeckel transcript , its translation would end at a premature stop codon shortly after the mutation site ( Figure 1C and Figure S3 ) . These mRNA molecules are more likely subjected to NMD , explaining the low abundance observed in SCKL2 cell lines . However , the observed symptoms can not be explained by a reduction in full-length CtIP transcript , as no changes in the expression were observed in the different cell lines . Therefore , we speculated that some of the truncated form of CtIP protein has to be expressed in those cells and that the expression of this aberrant form of CtIP is causing the disease . To test this hypothesis , we analyzed CtIP in protein samples from lymphoblasts obtained from a SCKL2 patient ( CtIPs/s ) , an asymptomatic family member ( CtIP+/s ) and an unrelated control individual ( CtIP+/+ ) . As shown in Figure 2A , no discernible changes in full-length CtIP protein levels were observed between the different cell lines when using an antibody raised against the CtIP C-terminus [17] , in agreement with no changes at the mRNA level and indicating that accumulation of full-length CtIP is not compromised in SCKL2 patients . Notably , however , when using an antibody raised against the CtIP N-terminus ( Anti CtIP-Nt ) [17] , an additional shorter protein species , undetectable in CtIP+/+ lymphoblasts , was observed in both CtIPs/s and CtIP+/s cell extracts ( Figure 2A ) . Unfortunately , this antibody seemed less specific than the one raised against the C-terminus . Thus , to confirm that this band truly represented a form of CtIP , we immunoprecipitated all forms of the protein from etoposide treated cells by using an antibody raised against the central part of the protein ( Sigma ) and blotting with a different CtIP antibody ( Novus Biological Ltd ) . Using this approach we confirmed that in protein samples from CtIPs/s and CtIP+/s lymphoblast a new , shorter species of CtIP of around 100 kDa appeared , whereas it was absent in the unrelated control protein samples ( Figure 2B ) . We therefore denoted this protein , CtIPSCKL2 . Quantifying the abundance of this polypeptide revealed that , despite the low abundance of the CtIPSeckel transcript in CtIPs/s cells , the CtIPSCKL2 protein is present in amounts comparable to full-length CtIP ( Figure 2C ) . An additional band between full-length CtIP and CtIPSCKL2 was also observed in all three extracts with all the tested antibodies ( Figure 2A and 2B , double asterisk ) . We hypothesize that this represents an additional form of CtIP , which is not related to the phenotypes we are studying here . Alternative spliced CtIP mRNA has been found previously and at the protein level at least one CtIP variant has been isolated ( Genbank Accession # NP_976037 ) . Full-length CtIP is hyper-phosphorylated upon DNA-damage in a BRCA1-dependent manner by the apical checkpoint kinase ATM [11] , [18] , [19] ( Figure 2A , top panel; for phosphorylation sites , see Figure 1D ) . Strikingly , although the truncated CtIPSCKL2 protein retains the site required for its interaction with BRCA1 ( Ser-327 ) [20] and contains all known sites for ATM-mediated phosphorylation in response to DNA damage ( Figure 1D ) [18] , [19] , it was devoid of detectable DNA-damage induced modification as assessed by changes in its electrophoretic mobility ( Figure 2A ) , suggesting that it represents a DNA-damage unresponsive form of the protein . Accordingly , immunoprecipitation of a full length and the truncated form of CtIP from etoposide treated lymphoblast using the CtIP antibody from Sigma and blotting with antibodies that recognize sites phosphorylated by ATM readily rendered a strong phosphorylation signal of full-length CtIP but no signal for the truncated protein ( Figure 2D ) . Because CtIP promotes DSB processing into RPA-coated ssDNA required for ATR activation [11] , we explored whether CtIPs/s cells were defective in this process . The RPA complex is phosphorylated once it is bound to ssDNA , with Ser-4/Ser-8 ( S4/S8 ) phosphorylation of the RPA32 subunit being a readout of DSB processing [11] , [12] , [21] . In line with our other findings , after treating cells with the topoisomerase inhibitor etoposide , we observed that CtIPs/s cells exhibited reduced RPA32 S4/S8 phosphorylation as compared with CtIP+/+ cells , with the difference being most evident at low etoposide doses ( Figure 2A ) . In accordance with these data and with SS being associated with defective ATR activation [2] , [5] , we also found that CtIPs/s and CtIP+/s cells exhibited a mild defect in ATR signaling as measured by Chk1 phosphorylation , with this again being more evident at low etoposide doses ( Figure 2A; see Figure 2E for quantification ) . Notably , cells from heterozygous CtIP+/s also displayed small impairments in RPA32 and Chk1 phosphorylation ( Figure 2A and 2E ) , suggesting that heterozygous carriers of the CtIPSCKL2 are mildly defective in ATR signaling at the cellular level . To explore things further , we analyzed a second readout for DNA-end processing: RPA focus formation in cells treated with etoposide . To avoid potential differences in numbers of damaged cells , we quantified cells displaying γH2AX foci – which are well-established markers for DSBs – and the proportion of these γH2AX positive cells that also displayed RPA foci . Notably , whereas the proportions of cells positive for γH2AX and RPA foci were similar in CtIP+/+ and CtIP+/s lymphoblasts , CtIPs/s cells exhibited a mild but significant reduction in etoposide-induced RPA-focus formation ( Figure 2E; see representative images on the left and quantifications on the right ) . This decrease did not reflect differences in cell cycle distributions ( Figure S4 ) or rates of cell-cycle progression because the amount of γH2AX positive cells was similar for all three genotypes ( Figure 2E ) . Similarly , when we assessed cells for readouts of spontaneous DNA damage , which most likely arises through replication forks encountering DNA lesions , we found that the proportion of γH2AX positive cells that also displayed RPA foci was lower for CtIPs/s lymphoblasts than for CtIP+/+ and CtIP+/s cells ( Figure 2G ) . Microcephaly , being a core manifestation of SS , is thought to result from reduced proliferative potential in the developing nervous system , most likely due to increased cell death of neuronal stem cells or progenitor cells in the rapidly expanding fetal brain [1] . As only lymphoblasts were available from SCKL2 patients , we could not analyze their DNA-damage sensitivity by traditional clonogenic assays . Instead , we cultured CtIPs/s , CtIP+/s and CtIP+/+ lymphoblast cells to the same density , treated them with etoposide or DMSO as a control and then , at 24 hour intervals , counted the number of total cells present together with the number of viable cells in the cultures . As shown in Figure 3A and 3B , when cells were mock-treated , they grew without a lag phase , with similar growth rates being exhibited by all three genotypes . After etoposide treatment , both the total number of cells and the number of viable cells dropped in all cases . Notably , in accordance with our other data , CtIP+/+ cells recovered faster than CtIPs/s cells , whereas CtIP+/s cells showed an intermediate response . These findings correlated with higher levels of apoptosis , measured by PARP cleavage , in CtIPs/s cells as compared to CtIP+/+ lymphoblasts , with CtIP+/s cells once again displaying an intermediate phenotype ( Figure 3C ) . These results thus provided indirect support for reduced proliferative potential being a factor in the pathogenesis of the SCKL2 neuro-developmental disorder . The mutated CtIPSCKL2 protein is estimated to comprise 782 amino acid residues , whereas full-length CtIP comprises 897 residues . Interestingly , while CtIPSCKL2 retains the CtIP dimerization domain and interaction sites for several of its protein partners , it lacks residues 790-897 that promote MRN binding and are required for DSB resection[11] . The region lost in CtIPSCKL2 also lacks Thr-847 , a key cyclin-dependent kinase ( CDK ) site that controls CtIP activity in response to DNA damage [12] , [22] , and lacks two small regions conserved between CtIP and its yeast counterparts Sae2 and Ctp1 ( Figure 1D ) [11] , [12] , [22] , [23] . These issues suggested to us that CtIPSCKL2 might not only be non-functional in the DDR but might also act in a dominant-negative manner in the context of the full-length CtIP protein that is also expressed in SCKL2 cells ( Figure 2A ) . To test this hypothesis , we expressed green-fluorescent protein ( GFP ) -tagged C-terminally truncated CtIP ( CtIPΔC ) protein [11] in cells wild-type for CtIP , and analyzed the impact of this on RPA focus formation after etoposide treatment . Significantly , over-expression of GFP-CtIPΔC but not of GFP-CtIP full-length or GFP alone hampered RPA-focus formation in primary human fibroblast and osteosarcoma cells ( Figure 4A and 4B ) . The clinical manifestation of SCKL2 is inherited recessively; that is , CtIP+/s family members , while showing mild cellular phenotypes that are intermediate between CtIPs/s and CtIP+/+ ( Figure 2 and Figure 3 ) , do not display overt clinical symptoms . We therefore hypothesized that the ratio between the dominant negative and full-length forms of CtIP might modulate the strength of the defect in RPA-focus formation and ATR signaling . To test this idea , we assessed RPA focus formation in human osteosarcoma cells that had been transfected with both GFP-CtIP and GFP-CtIPΔC in varying ratios ( Figure 4C; ratios of transfected DNA are shown below the plot and the percentage of CtIP that is full-length is stated below the western blot ) . Strikingly , when more wild-type CtIP than CtIPΔC was transfected , no effect on RPA focus formation was observed . However , once GFP-CtIPΔC reached over 70% of total CtIP , a moderate but significant reduction in RPA focus formation was observed after etoposide treatment , with more substantial defects in focus formation being observed at yet higher GFP-CtIPΔC levels ( Figure 4C ) . Accordingly , Chk1 phosphorylation on Ser-345 was also impaired once the ratio between GFP-CtIPΔC and CtIP full-length was above 70∶30 ( Figure 4D ) . These data therefore supported a model in which the C-terminally truncated CtIP protein serves as a dose-dependent dominant-negative mutant . During the course of the above experiments , we observed that , despite using increasing ratios of GFP-CtIPΔC DNA , the amount of CtIPΔC protein rapidly saturated and was always considerably higher than the full-length CtIP protein expressed from similar amounts of plasmid . This therefore mimicked the situation in cells from SCKL2 patients , in which low amounts of the CtIPSeckel transcript produced substantial amounts of CtIPSCKL2 protein . Notably , previous work has shown that CtIP is subject to a rapid turnover mediated by the E3 Ubiquitin ligase SIAH1 [24] , with the interaction with SIAH1 being mapped to the CtIP C-terminus [24] . To analyze the turnover of CtIPΔC and full-length CtIP , we transfected HEK293 cells with GFP-CtIPΔC and FLAG-CtIP . Although we were unable to co-immunoprecipitate SIAH1 either with full-length or truncated CtIP ( data not shown ) , we observed that the full-length ( FLAG ) but not the truncated form ( GFP ) of CtIP was stabilized upon addition of the proteosome inhibitor MG132 ( Figure 5A ) . Along the same lines , inhibition of protein synthesis by cycloheximide had a much more pronounced effect on the levels of full-length CtIP than on CtIPΔC ( Figure 5B ) . Collectively , these data suggested that the pathological form of CtIP , CtIPSCKL2 , is not subject to normal proteosome dependent turnover , thereby explaining how small amounts of an alternative spliced transcript can produce enough protein to generate a dominant negative effect . CtIP activity relies on various protein-protein interactions , including its homodimerization and interaction with MRN , BRCA1 , RB and PCNA . We explored the possibility that truncated CtIP can sequester some or all of these partners . First , we analyzed the ability of different forms of GFP-tagged CtIP to be recruited to sites of DNA damage in the absence of endogenous CtIP . To do so , we generated cell lines that stably expressed siRNA-resistant GFP-tagged versions of full-length and truncated CtIP , and then treated these cells with an siRNA oligonucleotide to specifically target the endogenous CtIP . As shown in Figure 5C , full-length CtIP and CtIPΔC were recruited with similar kinetics to sites of microirradiation ( quantification in Figure 5D; also see Videos S1 and S2 ) . Similar results were obtained when endogenous CtIP was present ( that is , without siRNA treatment; data not shown ) . Moreover , the residence times at sites of DNA damage were almost identical for full-length CtIP and CtIPΔC ( Figure 5E ) . These data supported a model in which CtIPΔC , previously shown to be non-functional , can localize to sites of DNA damage but is then unable to effectively promote ATR signaling and DNA repair . One possibility is that , as it still retains an intact homo-dimerization domain , the truncated form of CtIP can form an inactive hetero-dimer with full-length CtIP . Consistent with this idea , immunoprecipitation of GFP-CtIPΔC from cells also expressing full-length FLAG-CtIP revealed that the two proteins did indeed interact ( Figure 5F , lane 1 ) . Strikingly , blocking the interaction of CtIPΔC with full-length CtIP by deleting the dimerization domain suppress its ability to act as a dominant negative , supporting an scenario in which CtIPΔC sequester full-length CtIP ( Figure 4B ) . Furthermore , GFP-CtIPΔC co-immunoprecipitated with the MRE11 subunit of the MRN complex , even in the absence of transfected full-length CtIP ( Figure 5F , lanes 1 and 2 ) . Collectively , these data supported the idea that the truncated form of CtIP acts in a dominant-negative manner to partially impair DSB processing and ATR activation .
We have defined point mutations in CtIP associated with the human congenital microcephaly syndromes , Seckel ( SCKL2 family ) and Jawad . Although the lack of biological samples from the Jawad family prevented us from analyzing the predicted appearance of the shorter species of CtIP mRNA and protein in this case , we hypothesize that in each disease , the causative mutation leads to a transcript with a premature stop codon , yielding a C-terminally truncated form of CtIP that partially hampers DSB resection and ATR activation . It is possible that in the CtIPj mutant , the shorter form of RNA will be subject to nonsense-mediated decay ( NMD ) and no truncated protein will be present . Although this may be the case , degradation of the mutated CtIPj transcript by NMD is unlikely to be complete , as it has been shown that CtIP null mutants are embryonic lethal in mice , at least in part due to their inability to overcome RB-mediated G1 arrest [25] . Thus , we speculate that , in CtIPj/j patients , in whom the only form of the CtIP transcript is the mutated one , some truncated protein will be present to allow the G1/S transition . In this regard , we note that , even if the transcript is largely degraded by NMD , the reduced turnover that we have observed for C-terminally truncated CtIP ( both the CtIPSCKL2 and GFP-CtIPΔC ) will allow enough protein to accumulate to perform CtIP's essential functions , while impairing DNA resection . Our results highlight similarities between CtIP-SCKL2 and ATR-SCKL1 . Notably , inactivation of either CtIP or ATR is embryonic lethal in mouse , and in both SCKL1 and SCKL2 , a point mutation generates an alternative pre-mRNA splice site that leads to the cellular and physiological phenotypes associated with SS . However , in SCKL1 , the aberrant mRNA form is not stable and the shorter protein has never been detected in cells [2] . In this case , the disease is caused by a reduction in the amount of full-length ATR mRNA transcript that leads to reduced ATR protein levels [2] . By contrast , in SCKL2 cells , no changes in the overall levels of full-length mRNA or CtIP protein are observed ( Figure 2A and data not shown ) . Instead , the low abundance , shorter version of the CtIP RNA is translated , producing a C-terminal CtIP truncation mutant that seems to be more stable than the wild-type protein , as much smaller amounts of mRNA are enough to produce similar protein levels . In agreement with this model , we have found that the levels of CtIPΔC protein are less affected by protein synthesis inhibition than full-length CtIP , and that levels of the full-length but not the truncated form of CtIP is induced by proteosome inhibition . Collectively , our data therefore indicate that , in the case of SCKL2 , it is the presence of this truncated CtIP protein and not the reduction of full-length protein that causes the cellular phenotypes . We hypothesize that the resulting inability of SCKL2 cells to respond optimally to endogenously-arising DNA damage lowers the apoptotic threshold in SCKL2 patient cells , reduces their proliferative potential and thus causes the developmental phenotypes observed in these patients . Based on our analyses , we conclude that in SCKL2 patients , the mutated , C-terminally truncated form of CtIP impairs DSB processing and the formation of RPA-coated ssDNA . This impaired ssDNA production thus partially hampers ATR activation and results in hypersensitivity to DNA damage and induction of apoptosis . Although most of our observations were made when we used external sources of DNA-damaging drugs , we also observed that SCKL2 cells exhibited reduced RPA DNA-damage staining when grown in unperturbed conditions , possible due to endogenously-arising DNA damage . It is noteworthy that the predicted C-terminal truncated forms of CtIP in SCKL2 and Jawad cells lack the normal CtIP C-terminus , which is a key regulatory part of the protein . First , this region bears an MRN interaction domain [11] that is crucial for DNA-end resection . Although a second MRN interaction point has been found in the very N-terminal part of the protein [26] , the C-terminal region is essential for CtIP-mediated activation of MRN-associated nuclease activity . This could explain how , although GFP-CtIPΔC still interacts with the MRN complex , it renders the complex unable to effectively perform DNA end resection [11] . Moreover , the homology regions between CtIP and its budding yeast and fission yeast counterparts Sae2 and Ctp1 , respectively [23] are also located at the C-terminal part of the CtIP protein that is affected by the SCKL2 and Jawad mutations . In fact , the Thr-847 CDK target site of CtIP , which is crucial for ssDNA formation , is also missing in CtIPSCKL2 and CtIPJawad [12] , [22] . Consistent with these issues , we have found that , unlike the wild-type protein , CtIPSCKL2 is not detectably phosphorylated in response to DNA damage , despite it being able to hetero-dimerize with the full-length CtIP protein and despite it retaining sites for interaction with proteins such as BRCA1 and RB . Consequently , we propose a model in which this DDR unresponsive form of CtIP sequesters the full-length CtIP protein and/or some of its interaction partners in unproductive complexes , thereby blocking effective DSB processing . In accord with such an idea , we have found that C-terminally truncated CtIP is actively recruited to DNA damage sites with apparently the same kinetics and residence time as full-length CtIP . Moreover , CtIPΔC species that are not able to interact with full-length CtIP due to the deletion of the dimerization domain lose the ability to act as a dominant negative , favouring this model . While we do not know why these complexes are inactive , an attractive model is that CtIP acts as priming endonuclease , as has been proposed for its budding yeast counterpart Sae2 [27] . In this scenario , an initial endonucleolytic attack of the DNA molecule by CtIP will create the substrate for the MRN complex . As many nucleases work as dimers , it is possible than a CtIP/CtIPSCKL2 dimers will be non functional , an idea supported by the fact that Sae2 mutants lacking the C terminal part , conserved in CtIP , have a partial impairment of its endonucleolytic activity [27] . Unfortunately , it has not yet been proved that CtIP can act as an endonuclease . In fact , CtIP - as Sae2 - lacks of any putative nuclease domain . Further biochemical work characterizing the activity or activities of CtIP is needed to clarify this point . Our findings also support a model in which the strength of the phenotype caused by CtIPSCKL2 depends on the ratio between the levels of full-length and truncated forms of CtIP , presumably because this ratio affects the levels of functional CtIP-containing complexes . Consequently , it seems that a threshold of this ratio must be surpassed in order to hamper ATR signaling sufficiently to yield a measurable phenotype . Interestingly , this threshold appears to be different for cellular and clinical symptoms because heterozygous SCKL2 carriers do not manifest clinical symptoms while they do present cellular phenotypes , albeit intermediate between homozygous and wild-type cells . Consequently , a dominant-negative form of CtIP causes SCKL2 Seckel syndrome recessively . To our knowledge this is the first example of such type of inheritance in man .
Splice site predictions were evaluated by submitting the SCKL2 CtIP sequence from genomic position 18835580 to 18840493 , spanning exon 15 to 16 , to the following online splice-site-prediction algorithms: http://www . cbs . dtu . dk/services/NetGene2/ http://www . fruitfly . org/seq_tools/splice . html For mutation detection , PCR was done on genomic DNA using intronic primers designed for amplification of all exons , including both UTR regions and an average 200 bp exon/intron overlap ( Table S1 ) . Cycle sequencing was performed directly on products . For mutation analysis , RNA was isolated from EBV transformed B-lymphoblastoid cell lines from SCKL2 family members and unrelated healthy control individuals using “Nucleospin Total RNA Isolation Kit II” according to manufacturer's instructions . RNA was eluted in 60 µl RNase-free H2O . RNA concentrations were measured by optical density and purity of the RNA was controlled by gel electrophoresis . cDNA was made using “iScriptTM Select cDNA Synthesis Kit” with mix of random hexamer and poly-dT primers using 1 µg of total RNA as synthesis template . Remaining procedures were according to manufacturer's protocols . For mutation-specific PCR , Roche Expand High Fidelity PCR System enzyme mix was used . The PCR solution was mixed up to a total volume of 20 µl consisting of: 2 µl cDNA elution ( diluted 1∶10 ) , 2 µl 10xPCR buffer ( Roche ) , 0 . 4 µl dNTP ( 0 . 2 mM final concentration ) 0 . 6 µl of each primer ( 0 . 3 µM final concentration ) ( forward primer: TGGTTAGTGAAACCGTTCTCTT and reverse primer: TGCAACTGAGAAGCCATAATTAAA ) , 0 . 1 µl Taq polymerase ( 5U/μl ) and various proportions of MgCl2 , DMSO and dH2O . Reactions consisted of 2 min at 94°C followed by 38 cycles of 30 sec at 94°C , 30 sec at 60°C and 1 . 5 minutes at 72°C , ending with 8 min at 72°C . The PCR products were purified by gel electrophoresis , extracted and directly sequenced . Transcript expression was examined by qRT-PCR using the LightCycler 480 Real-Time PCR System ( Roche diagnostics ) and the DNA-binding dye SYBR Green ( Invitrogen Corporation , Carlsbad , California , USA ) . For CtIPwt and CtIPSeckel comparison within the same cell lines , 2 primer-pairs were designed to specifically target each of the transcripts ( Primer sequences are available upon request ) and were run in parallel with 2 in-house validated normalizers . Expression levels between cell lines were examined using one target primer-pair for each transcript normalized to above-mentioned normalizers . cDNA from two independent iScript ( Bio-Rad Laboratories , Inc . ) reactions were analyzed for all samples which were run in triplets . SYBR Green amplification mixtures ( 10 µl reactions ) contained SYBR Green master mix , 0 , 5 µM of each forward and reverse primer , and 2 µl of template cDNA . The PCR cycling conditions were as follows: 10′ at 95°C , followed by 40 cycles of 10″ at 95°C , 20″ at 60°C and 30″ at 72°C . After PCR amplification , a melting curve was generated for every PCR product to check the specificity of the PCR reaction ( absence of primer dimers or other nonspecific amplification products ) . Each assay included a no-template control . The threshold cycle ( Ct ) values of LightCycler 480 Software , Version 1 . 5 ( Roche diagnostics ) were exported to Excel ( Microsoft Corp . , Seattle , Washington , USA ) for relative quantification analysis using a modified delta-delta Ct method with efficiency correction . EBV transformed lymphoblasts were grown in RPMI buffer containing 15% fetal bovine serum ( BioSera ) , 100 units/ml of penicillin , and 100 µg/ml of streptomycin ( Sigma-Aldrich ) . BJ and U2OS cells were grown in DMEM ( Sigma-Aldrich ) supplemented with 10% fetal bovine serum , 100 units/ml of penicillin , and 100 µg/ml of streptomycin . Transient transfection of GFP-CtIP , GFP-CtIPΔC and GFP was with a microporator ( MicroPorator ) following the manufacturer's protocols . Experiments were performed 48 h after transfection . Extracts were prepared in Laemmli buffer ( 4% SDS , 20% glycerol , 120 mM Tris-HCl pH 6 . 8 ) , proteins were resolved by SDS-PAGE and transferred to nitrocellulose followed by immunoblotting . R . Baer ( Columbia University ) provided the mouse monoclonal antibodies raised against the CtIP C-terminus or N-terminus . Other antibodies were from Novus Biological ( CtIP ) , Sigma ( CtIP , Tubulin ) , Abcam ( RPA32 ) , Bethyl Laboratories ( RPA32-pS4/S8 ) , Cell Signaling Technology ( PARP , phospho-S345 Chk1 ) and Santa Cruz ( Chk1 ) . Western quantification was performed using additional blots , equally loaded to the ones shown in the figures and scanned with a LI-COR Odyssey infrared imaging system . Cells were transfected with FLAG-CtIP and GFP-CtIPΔC . Two days after transfection , cells were treated with 10 µM MG132 ( Sigma ) or 150 µg/ml cycloheximide ( Sigma ) . At the indicated times , cells were washed with ice cold PBS and collected in Laemmli buffer . Proteins were resolved by SDS-PAGE and blotted with an anti-Flag ( Sigma ) or anti-GFP ( Roche ) antibody as indicated . Cells transfected with GFP-CtIPΔC or GFP and either FLAG-CtIP or FLAG , were grown for two days and then collected in RIPA buffer ( 50 mM Tris-HCl , pH 7 . 4 , 1% NP-40 , 0 . 25% Na-deoxycholate , 150 mM NaCl , 1 mM EDTA and 0 . 1% SDS ) . GFP immunoprecipitation was performed by using a GFP-Trap ( Chromotek ) , following the manufacturer's instructions . After protein electrophoresis and transfer to nitrocellulose filters , membranes were blotted with anti-GFP ( Roche ) , anti-FLAG ( Sigma ) or anti-Mre11 ( Novus Biological ) . Cells were colleted by centrifugation , washed twice with ice cold PBS and resuspended in RIPA buffer ( 50 mM Tris-HCl , pH 7 . 4 , 1% NP-40 , 0 . 25% Na-deoxycholate , 150 mM NaCl , 1 mM EDTA and 0 . 1% SDS ) . Immunoprecipitation was performed by using a CtIP antibody from Novus . After protein electrophoresis and transfer to nitrocellulose filters , membranes were blotted with anti-CtIP from Sigma or anti-phospho S/TQ ( Novus Biological ) . For RPA focus detection , lymphoblasts were treated with 10 µM of etoposide or DMSO . One hour afterwards , cells were washed twice with PBS and deposited on coverslips using a CytoSpin funnel on a CytoCentrifuge . Following pre-extraction for 5 min on ice ( 25 mM Hepes pH 7 . 4 , 50 mM NaCl , 1 mM EDTA , 3 mM MgCl2 , 300 mM sucrose and 0 . 5% Triton X-100 ) , cells were fixed with 4% paraformaldehyde ( w/v ) in PBS for 15 min . Cover-slips were washed three times with PBS and then co-immunostained with antibodies against γH2AX ( Cell Signaling Technology ) and RPA32 ( Lab Vision ) . For detection , Alexa Fluor-594 ( red ) and -488 ( green ) conjugated secondary antibodies were used ( Molecular Probes , Paisley , UK ) . Samples were visualized with an Olympus upright confocal microscope by sequential scanning of the emission channels . U2OS and BJ cells transfected with GFP fusions were grown on cover-slips for two days after transfection , treated with etoposide as described and fixed and co-immunostained as above , with the exception that Alexa Fluor-594 ( red ) and -647 ( far red ) conjugated secondary antibodies were used . Localized DNA damage was generated by exposing cells to a UV-A laser [28] , [29] . Cells plated on glass-bottomed dishes ( Willco Wells ) and were pre-sensitized with 10 mM 5-bromo-20 deoxyuridine ( BrdU , Sigma-Aldrich ) in phenol red-free medium ( Invitrogen ) for 24 h at 37°C . Micro-irradiation was with a FluoView 1000 confocal microscope ( Olympus ) equipped with a 37°C heating stage ( Ibidi ) and a 405 nm laser diode ( 6 mW ) focused through a 60X UPlanSApo/1 . 35 oil objective to yield a spot size of 0 . 5 to 1 mm . Time of cell exposure to the laser beam was ∼250 ms ( fast-scanning mode ) . Laser settings ( 0 . 40 mW output , 50 scans , SIM scanner ) were chosen to generate a detectable DDR restricted to the laser path in a pre-sensitization-dependent manner without detectable cytotoxic effects . FRAP analyses were performed on the microscope used for laser micro-irradiation when the accumulation of the GFP-tagged protein on the laser track reached its maximal steady-state level . After a series of three pre-bleach images , a rectangular region placed over the laser-damaged line was subjected to a bleach pulse ( five scans with 488 nm argon laser focused through a 360 UPlanSApo/1 . 35 oil objective , main scanner , 100% AOTF acousto-optical tunable filter , slow scanning mode ) , followed by image acquisition at fastest speed . Average fluorescent intensities in the bleached region were normalized against intensities in an undamaged nucleus in the same field after background subtraction to correct for overall bleaching of the GFP signal due to repetitive imaging . For mathematical modelling of GFP-tagged protein mobility , ( It - I0 ) /Ipre values were plotted as a function of time , where I0 is the fluorescence intensity immediately after bleaching and Ipre is the average of the three pre-bleach measurements . Estimation of mobile protein fraction ( A ) and residence time ( t ) were performed using Prism 4 software assuming the existence of one protein population using the following equation: y ( t ) = A ( 1 - exp ( -t/t ) ) . To determine cell-cycle distributions , cells were fixed with 70% ethanol , incubated for 30 min with RNase A ( 250 µg/ml ) and propidium iodide ( 10 µg/ml ) at 37°C . For each experiment 104 cells were analyzed for fluorescence and recorded by a FACSAria flow cytometer ( Becton Dickinson , Mountain View , CA ) ; cell debris was excluded on the basis of forward and side light-scattering properties . Cell cycle distribution was determined from DNA fluorescence histograms using the CellFit software ( Becton Dickinson ) . Cells were grown to ∼50 , 000 cells/ml , and then split in two . Half of the culture was treated with 10 µM of etoposide , and the rest was mock treated with DMSO . A sample was taken every 24 h for 10 days to determine cell number/viability . Cell numbers were determined by direct counting with a Beckman Coulter Cell and particle counter . Viable cell were measured with a Cell Counting Kit-8 ( Dojindo Lab . , Kumamoto , Japan ) following the manufacturer's instructions .
|
Cellular DNA is frequently damaged through the actions of exogenous and endogenously arising DNA damaging agents . To maintain genome integrity , cells have evolved complex mechanisms to detect DNA damage , signal its presence , and mediate its repair . The importance of such mechanisms is evident because inherited defects in them can cause embryonic lethality or severe genetically inherited diseases . The clinical manifestations of such diseases are complex and include growth delay , mental retardation , skeletal abnormalities , and predisposition to cancer . While most such syndromes are inherited recessively , in some cases they are inherited dominantly . Here , we show that mutations in CtIP/RBBP8 cause related disorders: Seckel and Jawad syndromes . In addition to revealing how mutated CtIP impairs responses to DNA damage in Seckel cells , we establish that , despite the recessive mode of inheritance for this syndrome , the Seckel mutation has a dominant manifestation at the cellular level . To our knowledge , this represents a new form of molecular mechanism for recessive inheritance of a human disease . Furthermore , the aberrantly spliced mRNA is expressed at very low levels and yet significantly impairs cellular functions and causes severe clinical symptoms . This should provide new awareness that even very subtle splice mutations may have pronounced pathogenic potential .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"chromosome",
"structure",
"and",
"function",
"dna",
"signaling",
"in",
"cellular",
"processes",
"chromosome",
"biology",
"biology",
"molecular",
"biology",
"signal",
"transduction",
"autosomal",
"recessive",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"dna",
"repair",
"human",
"genetics",
"molecular",
"cell",
"biology",
"genetics",
"of",
"disease",
"genetics",
"and",
"genomics",
"signaling",
"cascades"
] |
2011
|
CtIP Mutations Cause Seckel and Jawad Syndromes
|
The smallest algae , less than 3 μm in diameter , are the most abundant eukaryotes of the World Ocean . Their feeding on planktonic bacteria of similar size is globally important but physically enigmatic . Tiny algal cells tightly packed with the voluminous chloroplasts , nucleus , and mitochondria appear to have insufficient organelle-free space for prey internalization . Here , we present the first direct observations of how the 1 . 3-μm algae , which are only 1 . 6 times bigger in diameter than their prey , hold individual Prochlorococcus cells in their open hemispheric cytostomes . We explain this semi-extracellular phagocytosis by the cell size limitation of the predatory alga , identified as the Braarudosphaera haptophyte with a nitrogen ( N2 ) –fixing endosymbiont . Because the observed semi-extracellular phagocytosis differs from all other types of protistan phagocytosis , we propose to name it “pomacytosis” ( from the Greek πώμα for “plug” ) .
In conventional phagocytosis , the caught prey is internalized , i . e . , enclosed by a phagocytic membrane inside the predator cell to form a food vacuole , within which prey is digested and its contents are absorbed through the vacuole membrane [1] . Apart from secure isolation of the prey from the environment , full closure of the food vacuole benefits the predator in a number of ways . The fully closed vacuole allows the predator to pump excess water to reduce the vacuole volume , to adjust pH inside the vacuole to facilitate prey digestion by lytic enzymes , and to contain lysed prey for efficient nutrient assimilation . Only refractory prey material , e . g . , moieties of cell wall , is egested when the closed food vacuole finally fuses back with the plasma membrane [2] . Thus , conventional phagocytosis of internalized prey requires enzymes , microfilament , microtubule , and membrane investments and can be limited by the predator size [3] . Phagocytosis of prey of similar size or bigger is difficult but achievable for protists . For example , some dinoflagellates use a feeding tube to inject lytic enzymes into prey and to extract digested prey contents [4 , 5] . Other dinoflagellates and several haptophytes form extracellular , yet closed , food vacuoles [6–8] . Such extensive extracellular vacuoles can only be completed by large predatory cells that can produce and stock sufficient amounts of the required investments . Compared to extracellular phagocytosis , internalization of similar-sized prey requires from the predator fewer investments but sufficient intracellular space free from organelles . In protists , the nucleus , mitochondria , and chloroplasts ( the latter in algae ) can vary in size , but these organelles cannot be smaller than a certain minimal volume . Owing to the presence of such “nonscalable” organelles [9] , the intracellular volume available for investment storage and prey internalization shrinks as a power function of the predator cell size . Consequently , small protists may be unable to internalize ( conventionally phagocytose ) similar-sized prey . To test that , we focused on feeding of the smallest algae ( <3 μm in diameter ) , whose chloroplast-packed cells in addition to the nucleus and mitochondria should have the minimal organelle-free space among free-living protists . According to our morphometric estimates , organelles occupy approximately 70% of a haptophyte alga with a cell volume of 2 . 8 ± 0 . 8 μm3 ( n = 10; S1 Fig ) . Even after taking into account scalable but vital cell components , e . g . , endoplasmic reticulum rich in ribosomes and enzymes , both the haptophyte alga as well as the smallest known prasinophyte alga with a cell volume of 1 . 1 to 5 . 7 μm3 [10] are still capable of internalizing a bacterial cell of 0 . 1 to 0 . 3 μm3 [11] . This is in agreement with the substantial indirect experimental evidence that , despite their diminutive size , the smallest ( 1- to 3-μm diameter ) algae are the main predators of bacterioplankton in the open ocean [12 , 13] . However , because of insufficient resolution of optical microscopy , phagocytosis by these algae could only be inferred [14] . In order to find out how algae less than 3 μm in size phagocytose similar-sized bacteria , we chose to study the smallest oceanic picoeukaryotic algae , plastidic eukaryote small ( PES ) , separated from other protists and bacteria living in seawater by flow cytometry . Using high-resolution electron microscopy to observe fine cellular details of the sorted algae , we found that their semi-extracellular bacterial phagocytosis—“pomacytosis”—differs from all other types of phagocytosis .
Low concentrations of bacterioplankton and PES ( 6 × 105 cells ml-1 and 4 × 102 cells ml-1 , respectively; S2 Fig ) in the studied region of the Eastern subtropical North Atlantic Ocean were typical for open ocean waters [11 , 13] . The main PES population was well defined by flow cytometry and selected for sorting ( S3 Fig ) . High-throughput barcoding analysis of flow-sorted PES cells ( S3 Fig ) yielded 10 , 416 high-quality ( ≥300 nt ) 16S rRNA gene reads and identified the dominant taxa: 51% of the amplicons were sequences of cyanobacteria , composed of Prochlorococcus ( 26% ) and unicellular diazotrophic cyanobacteria group A ( UCYN-A; 25% ) , and 38% of the amplicons were chloroplast sequences , the majority of which ( 58% ) belonged to the Braarudosphaeraceae , a coccolithophore family of the Haptophyta ( Fig 1 ) . The remaining chloroplast sequences belonged to 10 other types of small algae , each of which represented only a minor fraction of the PES cells ( Fig 1 ) . The negligible number of sequences of SAR11 alphaproteobacteria ( Rickettsidae ) —the most abundant bacteria in the samples ( thus the most probable by-sorted cells ) —validated the high purity of PES sorting . Analyses of nearly full-length ribosomal gene sequences confirmed the phylogenetic affiliation obtained with shorter amplicons . Full-length sequences of the 16S rRNA gene of Prochlorococcus and UCYN-A were 99% identical to high light–adapted Prochlorococcus marinus strain MIT9301 and 100% identical to the Candidatus Atelocyanobacterium thalassa isolate a long-term oligotrophic habitat assessment ( ALOHA ) [15] , respectively . The 18S rRNA gene sequence was 99% identical to a calcifying Braarudosphaera bigelowii isolate TMRscBb7 [16] ( S4 Fig ) and to a small noncalcifying alga collected from oligotrophic waters of the South East Pacific Ocean ( S4 Fig ) [17] , confirming the chloroplast 16S rRNA gene-based identification . Scanning and transmission electron microscopy ( SEM and TEM , respectively ) showed no curved rod-shaped cells of the most abundant SAR11 bacteria ( S5A Fig ) among flow-sorted PES cells . The absence of by-sorted SAR11 bacteria reaffirmed the high sorting purity . The majority ( 95% ) of the imaged PES cells ( 185 out of 195 ) were ball-shaped small cells with an estimated diameter of 1 . 3 ± 0 . 22 μm ( n = 33 , size corrected for 30% linear cell shrinkage during sample dehydration [18] ) . Some of them bore organic , noncalcified scales ( S6 Fig ) . These morphotypes represent coccolythophore life cycle stages found in nutrient-poor waters [19–21] . Among the sorted PES cells , there were no cells with external mineral investments , i . e . , pentagonal-shape liths characteristic of Braarudosphaera species found in nutrient-replete waters [22] . A few morphologically different cells ( 10 out of 195 examined cells ) had one or two well-preserved flagella ( S7 Fig ) that ruled out the artificial loss of external investments by the dominant alga . Out of 185 cells of the dominant alga , 155 ( 84% ) were associated with smaller coccoid cells 0 . 81 ± 0 . 08 μm ( n = 10 , size corrected for 30% linear cell shrinkage during cell dehydration ) in diameter ( Fig 2 ) . An additional intracellular body of the dominant algal cells was observed using TEM ( Fig 2E ) . The 0 . 47 ± 0 . 05 μm ( n = 4 ) diameter body occupied a particular location at the cell periphery next to one of the two chloroplasts . When the body was absent , a rupture in the algal cell wall was observed ( Fig 2F , thick arrow; S5B Fig ) , confirming that the body was intracellular but could be lost under mechanical stress caused by sorting PES cells directly on TEM grids . A similar intracellular “spheroid body” in B . bigelowii isolate TMRscBb7 was identified as an obligate N2-fixing UCYN-A endosymbiont [16]—cyanobiont . Contrary to the cyanobiont , the molecularly identified Prochlorococcus associated with PES is a free-living planktonic cyanobacterium that was numerous in the studied seawater ( 1 . 7 × 105 cells ml-1 ) . Synthesising the above evidence , we concluded that the UCYN-A amplicon derived from the “intracellular body” and the Prochlorococcus amplicons represented the extracellular cocci attached to the PES cells . We interpreted the latter association as phagocytosis of Prochlorococcus by the naked haptophyte ( hereafter referred to as B . bigelowii JC142 ) . We assigned the observed B . bigelowii cells to two major groups and one minor group , as follows: ( a ) alga with an associated Prochlorococcus , of which less than 50% cell surface is inside the cytostome ( 49% ) ; ( b ) alga with an associated Prochlorococcus , of which more than 50% cell surface is inside the cytostome ( 35% ) ; and ( c ) alga with a cytostome but without prey ( 16% ) ( Fig 2 ) . The cytostome is most likely used for shape- ( and possibly surface- ) selective prey recognition and capture . In support of the notion of selection , neither molecular nor microscopic evidence suggested that B . bigelowii JC142 fed on SAR11 alphaproteobacteria—the most abundant free-living bacteria in the studied seawater ( 2 . 8 × 105 cells ml-1; S2 Fig ) . The algae preferred to feed on less abundant Prochlorococcus ( 1 . 7 × 105 cells ml-1 ) , which made up only 27% of total bacterioplankton in the seawater ( 6 × 105 cells ml-1; S2 Fig ) , i . e . , the haptophyte selected on average one out of four encountered free-living bacterial cells . Because of high-purity PES sorting , the individual Prochlorococcus observed by SEM were not by-sorted cells but were in fact cells detached from the haptophytes during sorting ( e . g . , Fig 2D ) . Both intact and doughnut-shaped , deformed Prochlorococcus cells were observed ( S5 Fig ) . The intact , spherical Prochlorococcus ( 14 observed cells ) were probably at the start of pomacytosis , while the doughnut-shaped Prochlorococcus with a central small spot of depressed surface area ( 23 cells ) were at the end of pomacytosis ( S5 Fig ) . Similarly , deformed Prochlorococcus cells were observed by SEM and TEM ( Fig 2L and 2M ) , affirming that the deformation was a result of pomacytosis rather than an artefact of SEM sample preparation . High-power TEM revealed that in the groups ( a ) and ( b ) , the prey Prochlorococcus cell is fitted into a semicircular cytostomic depression , which—according to SEM—is in reality hemispherical and is anchored in the cytostome between the two algal chloroplasts ( Fig 2C and 2E ) . In all 155 specimens observed with the Prochlorococcus cell , the latter remains at least partially free of the algal cytostome membrane ( Fig 2 , S8 Fig ) . To our knowledge , this is the first observation of semi-extracellular phagocytosis of prey by a protist using a partially opened cytostome .
The B . bigelowii JC142 is the smallest haptophyte that was directly observed to phagocytose free-living bacteria . However , B . bigelowii ability to internalize the selected bacterium is evolutionary evidenced—its cyanobiont is of a phagocytic origin . The intracellular UCYN-A symbiont cell in B . bigelowii isolate TMRscBb7 is surrounded by a food vacuole membrane [16] . The presence of the UCYN-A cyanobiont further reduces the intracellular space of B . bigelowii available for prey internalization . The size of the UCYN-A symbiont of B . bigelowii JC142 is at the lowest end of the reported UCYN-A size range [15 , 16 , 23 , 24] . The cyanobiont occupies less than 5% of the estimated volume of the B . bigelowii JC142 cell , while the Prochlorococcus prey measures more than 20% of the algal volume . Perhaps the choice between conventional phagocytosis and pomacytosis depends on the size ratio between the alga and its prey . In order to overcome its space limitation , 1 . 3-μm B . bigelowii JC142 cell , pomacytoses 0 . 8-μm Prochlorococcus instead of using whole-cell phagocytosis . Selective feeding of B . bilgelowii JC142 on Prochlorococcus implies that , despite the internal supply of fixed nitrogen by the UCYN-A cyanobiont—or perhaps owing to this supply as well as to metabolic demands of the symbiont—the haptophyte could be limited in other main inorganic nutrients [25] , e . g . , phosphorus and iron . However , this limitation is unlikely because B . bigelowii JC142 was collected in the Eastern subtropical North Atlantic Ocean fertilised by aeolian dust from the Saharan desert . Consequently , the surface waters in the studied area are enriched in phosphate and iron [26] but are poor in nitrogen salts [27]—the environment that facilitates the growth of N2-fixing photoautotrophs . Instead of photoautotrophy , B . bigelowii JC142 cells unconstrained by inorganic nutrients , including nitrogen ( fixed by its cyanobiont ) , pomacytose Prochlorococcus . Therefore , the main nutrient the haptophytes gain from Prochlorococcus prey is , perhaps , fixed carbon . B . bigelowii may require fixed carbon because it has the cyanobiont . The UCYN-A cyanobiont lost its photosystem II complex ( PSII ) but retained its photosystem I ( PSI ) [28] to use light energy to fix N2 . In return for the shared fixed nitrogen , the Braarudosphaera host should share its fixed carbon with the cyanobiont [15 , 29] . Furthermore , to minimize inhibition of the cyanobiont N2 fixation , a B . bigelowii cell needs to keep its intracellularly dissolved O2 concentration low . Large host cells , e . g . , Rhizosolenia and Rhopalodia diatoms , do that by spatially segregating their chloroplasts from N2-fixing cyanobionts within their cells [30] . In the 1 . 3-μm B . bigelowii JC142 cell ( Fig 2E ) , O2 produced by the adjacent chloroplast could directly inhibit N2 fixation by the cyanobiont , and the haptophyte needs to reduce [29] if not to halt photosynthesis by its own chloroplasts . Consequently , both the host and cyanobiont become starved of fixed carbon and require its alternative , external source . In order to acquire that fixed carbon , B . bigelowii JC142 selectively pomacytose free-living Prochlorococcus cyanobacteria . Based on our observations ( Figs 1 and 2 ) , we suggest interpreting the reported association between the “unknown structure” and UCYN-A–bearing haptophyte ( Fig 6 in [29] ) as Prochlorococcus cell pomacytosed by the haptophyte . Low CO2 fixation by the haptophyte chloroplasts compared with high CO2 fixation by the “unknown structure”—Prochlorococcus ( Fig 6 in [29] ) —supports our suggestion that the Braarudosphaera could acquire fixed carbon from its prey rather than from its own chloroplasts . Perhaps , because a CO2-fixing Prochlorococcus cell also produces O2 , the B . bigelowii JC142 cell does not internalize it . Instead , live Prochlorococcus is kept segregated from the O2-sensitive cyanobiont , and the haptophyte keeps the cytostome semi-open to allow O2 dissipation . Therefore , pomacytosed Prochlorococcus could be viewed as a temporary chloroplast substitute . Conventional phagocytosis is a relatively quick process that usually takes seconds ( e . g . , [8] ) , and one seldom observes a protist predator in the process of internalizing prey . Because the majority of the B . bigelowii JC142 collected during six-hour sampling was in a process of feeding ( 84% held prey ) , pomacytosis should be a slow process that takes hours . The absence of internalized Prochlorococcus cells and nearly 1:1 ratio between pomacytosed Prochlorococcus with more than half cell surface exposed ( group [a] ) and with less than half cell surface exposed ( group [b] ) suggest that the haptophyte controls exposure of the prey cell to seawater . During slow pomacytosis , the predator could gain extra benefit from the prey that fixes CO2 and takes up nutrients through the cell wall exposed to seawater ( Fig 2E–2K ) . Unlike conventionally phagocyting cells , the pomacyting B . bigelowii JC142 detained Prochlorococcus in their cytostome without full internalization , perhaps; harvested fixed carbon released by prey; and egested the deformed , spent prey without full digestion ( S5 and S8 Figs ) . Therefore , a combination of intracellular space limitation ( primarily ) and physiological requirements of the tiny predatory alga ( secondarily ) leads to semi-extracellular phagocytosis of selected prey .
This is an oceanographic study carried out in the international waters . This research does not require special permission . The study was carried out in the Eastern subtropical North Atlantic Ocean ( 23° 37′ N 20° 43′ W ) on board the Royal Research Ship “James Cook” during the research cruise JC142 from November to December 2016 . Seawater samples from 25 m ( a representative depth of the surface mixed layer ) were collected using a rosette of 20-l Niskin bottles mounted on a conductivity-temperature-depth ( CTD ) profiler . All plastic- and glass-ware for handling seawater was prewashed with 10% HCl and rinsed with sampled seawater . Concentrations of total bacterioplankton , including Prochlorococcus and SAR11 , the latter as a population of cells with low nucleic acid content [31] , were determined by flow cytometry . Routinely , samples were fixed with 1% ( w/v ) paraformaldehyde ( PFA ) final concentration , stained with SYBR Green I DNA dye [11 , 32] , and analysed with the custom-modified FACSort instrument ( Becton Dickinson , Oxford , UK ) equipped with the blue diode laser ( 488 nm , 50 mW; Quantum Analysis , Munster , Germany ) using the CellQuest software . For determining concentrations of PES and Synechococcus and for cross-referencing microbial populations in the concentrated samples ( used for flow sorting ) , seawater samples were fixed with 2% PFA , stained with 0 . 1 μg ml-1 Hoechst 33342 ( final concentration ) , and analysed with the custom-built MoFlo XDP instrument ( Beckman-Coulter , High Wycombe , UK ) ( S2 Fig ) using the Summit 5 . 4 software . The first UV diode laser ( 355 nm , 100 mW; JDSU , CY355-100 , Thailand ) and the second blue diode laser ( 488 nm , 240 mW; Cobolt , Solna , Sweden ) were aligned through the first and third pinhole , respectively . Shallow angle light scatter ( forward scatter [FSC] ) of the UV light was detected using the 351 ± 5–nm optical filter and the H957-18 photomultiplier ( Hamamatsu , Japan ) . More sensitive H957-27 photomultipliers ( Hamamatsu ) were used for detecting particle fluorescence at four wavelengths ( 457 ± 25 nm , 530 ± 20 nm , 580 ± 15 nm , >643 nm ) and the three wavelengths ( 505–550 nm , 580 ± 15 nm , 670 ± 15 nm ) excited by the first and second laser , respectively . A reference mixture of yellow-green ( 505/515 nm ) 0 . 5-μm beads ( Life Technologies , Eugene , Oregon , US ) and multifluorescence 1 . 0-μm beads ( Fluoresbrite Microparticles , Polysciences , Warrington , Pennsylvania , US ) were used as an internal standard for both fluorescence and flow rates . The absolute concentration of beads in the stock solution was determined using syringe pump flow cytometry [33] . For flow sorting , microbes were gravity concentrated approximately 103-fold using sterile 0 . 2-μm pore size Sterivex filter units ( Millipore , Watford , UK ) attached directly to Niskin bottles . For molecular identification , concentrated microbial samples were fixed with Lugol iodine solution [34] and stored at +4°C before being flow sorted within 48 hours . Samples were discoloured with thiosulfate [34] and stained with Hoechst 33342 prior to sorting . For electron microscopy analyses concentrated samples were fixed with 2% PFA and stained with Hoechst 33342 prior to sorting . The same dominant distinct population of the smallest picoeukaryotic algae—PES—was flow sorted with the MoFlo XDP instrument ( S3 Fig ) using the Summit 5 . 4 software . The instrument was optically aligned , and its sorting purity and recovery were optimised using blue ( 350/440 nm ) 1 . 0-μm beads ( Life Technologies ) . Only PES cells gated by both gates ( S3B , S3C , S3E and S3F Fig ) were sorted . Purity of sorted PES cells was validated by the molecular and electron microscopy analyses . For TEM analyses , 1 × 103 to 2 × 103 target PES cells were flow sorted directly on formvar/carbon–covered 200 mesh copper grids ( Agar Scientific , Stansted , UK ) stained with 2% w/v Gadolinium ( aqueous solution ) , rinsed with pure deionized water , and stored in a desiccator for analysis ashore . The grids were examined at 200 keV with the Jeol 2011 LaB6 TEM instrument fitted with a Gatan UltraScan 1000 camera at the University of Warwick’s Research Technology Platform in Advanced Bioimaging in the United Kingdom . For SEM analyses , 20 × 103 target cells were flow sorted into sterile 1 . 5-ml microcentrifuge tubes containing aqueous solution of 1% glutaraldehyde ( Electron Microscopy Sciences ) . The tubes were stored at 4°C and brought ashore . The sorted cells were collected onto 0 . 2-μm pore size , 13-mm polycarbonate filters under low vacuum , dehydrated in the ethanol series , and critical point dried using 99 . 9% hexamethyldisilazane ( Sigma-Aldrich ) . The dehydrated filters were stored in a desiccator at room temperature . Prior to SEM analyses , the filters were sputtered with Au/Pd ( 3:2 ) to a thickness of 10 nm using the High-Resolution ( 208hr ) Sputter Coater coupled with the MTM20 film thickness controller ( Cressington ) . The filters were examined with the high-resolution SEM UltraPlus instrument ( Zeiss Gemini ) at 5 keV using the secondary electron detector at the Imaging and Analysis Centre of the Natural History Museum in London , UK . Cell dimensions were measured on both TEM and SEM micrographs using the ImageJ software [35] . The values obtained from the SEM micrographs were corrected to account for approximately 30% cell shrinkage [18] . Average cell volumes were calculated assuming a ball or spheroid shape of algal cells ( 4/3πa2b ) , a spherical segment for chloroplasts ( πh2[b-1/3h] ) , an ellipsoid for a nucleus ( 4/3π[a-h]2b ) , and half of this ellipsoid for a mitochondrion ( S1 Fig ) . For molecular analyses , 20 × 103 to 50 × 103 PES cells were flow sorted into sterile 1 . 5-ml microcentrifuge tubes . An aliquot of 2 μl containing approximately 2 × 103 cells was added into a 0 . 2-ml PCR tube containing 30 μl of Q5 High Fidelity Master Mix ( New England BioLabs ) complemented with primers and nuclease-free water ( Ambion ) . For full-length 16S or 18S rRNA gene amplification , we used 27f/1492r [36] or 63f/1818r [37] primers with annealing temperature of 59°C . The amplicons were added with A-tails ( OneTaq DNA polymerase , New England BioLabs ) , ligated to the pGEM T-Easy vector ( Promega ) , and transformed into the NEB 5-alpha competent Escherichia coli cells ( New England BioLabs ) . Plasmids from the positive colonies were sequenced with T7 and SP6 primers to cover the full amplicon length . The 18S rRNA gene sequences were aligned with 18 reference sequences of haptophytes ( 1 , 400 positions ) , and phylogenetic relationships for the dataset were calculated with MrBayes software [38] . For a massively parallel sequencing , hyper variable regions V3–V4 ( 490 bp ) were amplified by PCR using S-D-Bact-0341-b-S-17 and S-D-Bact-0785-a-A-21 primers [39] . The forward primer included the PGM barcode adapter ( Ion Xpres Barcode Adapters 1–96 Kit , ThermoFisher Scientific ) , and both primers were tailed with the Ion Torrent sequencing adapters to allow direct downstream multiplexed sequencing . Following amplification , PCR products of approximately 490 bp were gel purified with NucleoSpin Gel and PCR Cleanup kit ( Macherey-Nagel ) , and 1 . 5 ng of the product was used for template preparation with the Ion Torrent OneTouch System ( ThermoFisher Scientific ) . The templates were sequenced on an Ion Torrent PGM sequencer ( ThermoFisher Scientific ) using the Hi-Q sequencing chemistry . After sequencing , the individual sequence reads were first quality trimmed using the Ion Torrent software suite and then further processed using the bioinformatics pipeline of the Silva NGS project [40] . This involved quality controls for sequence length ( ≥300 bp ) and the presence of ambiguities ( <2% ) and homopolymers ( <2% ) . The remaining reads were split into individual sample FASTA files using mothur [41] and aligned against the SSU rRNA seed of the SILVA database release 119 . The classification was done by a local BLAST search against the SILVA SSU Ref 115 nonredundant ( NR ) database using BLAST 2 . 2 . 22+ with standard settings . The analysis gave ( semi ) quantitative information ( number of individual reads representing in a taxonomic pool ) on the composition of the original PCR amplicon pool [39] . The classification of plastidic SSU rRNA sequence reads was done by nucleotide BLAST search against the NR database at the National Center for Biotechnology Information ( NCBI; www . ncbi . nlm . nih . gov ) .
|
The global significance of microorganisms is a consequence of their astronomical numbers . This is certainly true for the smallest planktonic algae on earth , which are less than 3 μm in diameter and the most numerous eukaryotic organisms of the oceans . Contrary to the general belief that algae use only sunlight and dissolved mineral nutrients to grow , these microscopic plants consume large numbers of bacteria . Their acting as mini-predators on bacteria of nearly their own size is hard to imagine . A tiny algal cell is cramped with organelles—such as nucleus , mitochondria , chloroplasts—and there is simply no space inside this cell to engulf a large bacterium in the usual manner . To find out how the 1 . 3-μm–diameter haptophyte algae feed , we studied them using high-resolution electron microscopy . We found that prey handling by the alga differs from all other types of cell feeding . We showed that this alga holds the 0 . 8-μm–diameter prey in the open cytostome ( cell mouth ) and that , from among planktonic bacteria , the alga apparently selects a ball-shaped Prochlorococcus ( an abundant cyanobacteria responsible for most of global photosynthesis ) that tightly fits into the open cytostome like a plug . Instead of full prey digestion , we observed that the alga leaves behind the doughnut-shaped carcass of the prey . We conclude that such unusual feeding , which we call “pomacytosis” , of this tiny predatory alga is caused primarily by the space limitation inside its cell .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"ecology",
"and",
"environmental",
"sciences",
"plant",
"cell",
"biology",
"cell",
"processes",
"chloroplasts",
"plant",
"science",
"aquatic",
"environments",
"microscopy",
"plants",
"cellular",
"structures",
"and",
"organelles",
"marine",
"environments",
"sea",
"water",
"research",
"and",
"analysis",
"methods",
"scanning",
"electron",
"microscopy",
"marine",
"and",
"aquatic",
"sciences",
"short",
"reports",
"phagocytosis",
"algae",
"ribosomes",
"community",
"ecology",
"biochemistry",
"rna",
"transmission",
"electron",
"microscopy",
"eukaryota",
"plant",
"cells",
"ribosomal",
"rna",
"cell",
"biology",
"ecology",
"nucleic",
"acids",
"predation",
"trophic",
"interactions",
"earth",
"sciences",
"electron",
"microscopy",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"non-coding",
"rna",
"organisms"
] |
2018
|
“Pomacytosis”—Semi-extracellular phagocytosis of cyanobacteria by the smallest marine algae
|
The alveolar compartment , the fundamental gas exchange unit in the lung , is critical for tissue oxygenation and viability . We explored hepatocyte growth factor ( HGF ) , a pleiotrophic cytokine that promotes epithelial proliferation , morphogenesis , migration , and resistance to apoptosis , as a candidate mediator of alveolar formation and regeneration . Mice deficient in the expression of the HGF receptor Met in lung epithelial cells demonstrated impaired airspace formation marked by a reduction in alveolar epithelial cell abundance and survival , truncation of the pulmonary vascular bed , and enhanced oxidative stress . Administration of recombinant HGF to tight-skin mice , an established genetic emphysema model , attenuated airspace enlargement and reduced oxidative stress . Repair in the TSK/+ mouse was punctuated by enhanced akt and stat3 activation . HGF treatment of an alveolar epithelial cell line not only induced proliferation and scattering of the cells but also conferred protection against staurosporine-induced apoptosis , properties critical for alveolar septation . HGF promoted cell survival was attenuated by akt inhibition . Primary alveolar epithelial cells treated with HGF showed improved survival and enhanced antioxidant production . In conclusion , using both loss-of-function and gain-of-function maneuvers , we show that HGF signaling is necessary for alveolar homeostasis in the developing lung and that augmentation of HGF signaling can improve airspace morphology in murine emphysema . Our studies converge on prosurvival signaling and antioxidant protection as critical pathways in HGF–mediated airspace maintenance or repair . These findings support the exploration of HGF signaling enhancement for diseases of the airspace .
One approach to identifying mediators of alveolar formation and regeneration in the mammalian lung is to delineate the elemental events that attend airspace formation and then systematically investigate candidate proteins that harbor a compatible signaling repertoire in animal or cellular model systems . From a developmental perspective , the eruption of alveolar septae from a primordial saccule in early postnatal murine life requires localized epithelial proliferation , migration and resistance to apoptosis [1] , [2] . Furthermore , epithelial morphogenesis must be accompanied by a microvasculature that permits efficient diffusion of gases from the airspace lumen to the systemic vascular bed . A candidate mediator of such events is hepatocyte growth factor ( HGF ) . HGF is a pleiotrophic cytokine that promotes epithelial proliferation , morphogenesis , migration and survival [3] , [4] . HGF is also known to induce angiogenesis and inhibit epithelial apoptosis . We sought to determine whether HGF is critical for alveolar formation and might have a therapeutic role in alveolar regeneration . The HGF/c-Met signaling pathway incorporates all of the features of a key alveolar survival factor [5] . HGF is expressed with its receptor ( c-Met or Met ) in the vertebrate lung parenchyma . Upon HGF binding , c-Met undergoes autophosphorylation which initiates the recruitment of a variety of downstream signal transduction molecules ( reviewed in [6] ) . In selective models , HGF signaling supports postpneumonectomy lung growth , induces branching morphogenesis and ameliorates inflammatory lung injury [7]–[9] . Unfortunately , these artificial models imperfectly approximate the physiologic events required for alveolar formation and maintenance . Moreover , since elevated local and systemic HGF levels in patients with lung injury correlate with disease severity and poor outcomes , whether HGF is a participant in airspace repair or a marker of ongoing injury is a subject of controversy [3] . When HGF is administered during neonatal hyperoxia or early in the course of bleomycin lung injury , some measure of protection against lung damage is observed . However , no studies have addressed whether HGF/c-Met signaling is 1 ) required for alveolar formation or 2 ) has therapeutic value in animals with established pathologic airspace enlargement . Our studies utilize both loss-of-function and gain-of-function strategies in the murine lung to investigate the function of this pathway in murine airspace formation and airspace repair . To delineate a developmental role , we present a postnatal interrogation of mice deficient in HGF signaling in the airspace epithelia . To investigate mechanisms of airspace repair , we use the TSK/+ mouse model of genetic emphysema , a convenient platform to evaluate strategies for airspace regeneration . In studies here , we both assess whether HGF infusion can improve airspace caliber in this model and evaluate downstream pathways engaged in the therapeutic response . The goal is to invoke candidate mechanisms for HGF-mediated airspace repair with possible broad therapeutic utility . In the present study , we find that mice deficient in Met expression in alveolar epithelial cells exhibit impaired airspace morphology accompanied by a reduced abundance and survival of alveolar type II cells . We also show a reduction of vascularization in the lung parenchyma of Met-deficient mice , suggesting an intimate morphogenic connection between the epithelial and endothelial compartments . Pharmacologic augmentation of HGF signaling in a murine model of emphysema reverses airspace enlargement in the adult lung and is marked by akt and stat3 activation . Finally , in whole cell assays using primary and immortalized alveolar epithelial cells , we establish that hepatocyte growth factor signaling promotes cell survival , induces proliferation and scattering of alveolar epithelial cells , confers protection against cell death via akt activation and mediates antioxidant production . These data 1 ) support a critical role for HGF signaling in alveolar development and regeneration and 2 ) implicate downstream prosurvival signaling as a contributor to the airspace maintenance and reparative effects of HGF . Importantly , the studies suggest that developmental strategies for airspace formation are recapitulated in reparative contexts .
In order to invoke a role of the HGF/c-Met pathway in alveolar morphogenesis , we first established that the ligand and receptor are expressed in alveoli . Using immunohistochemistry , we show that c-Met is expressed in alveolar type II cells in the lungs of two- week old C57Bl/6 mice ( Figure 1A ) . We also show that the HGF ligand is expressed diffusely in the interstitium of the alveolar septum . HGF is notably excluded from alveolar epithelial cells , consistent with its known localization and deposition in other tissues ( Figure 1A and Figure S1A ) . Coimmunostaining for an alveolar epithelial marker ( SPC ) and c-Met in 2 week old lungs shows that the sites of c-Met expression are alveolar epithelial cells , airway epithelial cells and a subset of alveolar macrophages ( Figure S1B ) . Having established that HGF and c-Met are expressed in the developing murine lung , we investigated whether this pathway was critical for alveolarization . Yamamoto recently showed that mice with an alveolar epithelial cell specific deletion of Met had impaired late embryonic lung development , implicating the HGF/c-Met pathway in late lung development [10] . However , a dedicated analysis of the postnatal phenotype was not pursued . We generated mice deficient in Met expression in alveolar epithelial type II cells ( AECII ) . Conditional alveolar epithelial cell specific deletion of Met was achieved by crossing the well characterized SPC-rtta;otet-Cre cassette provided by Dr . Jeffrey Whitsett into Metf/f mice that harbor an inactivating conditional deletion in exon 16 provided by Dr . Snorri Thorgeirsson [11] , [12] . These tritransgenic mice ( SPC-rtta/+;otet-Cre;Metf/f , termed SPCMetf/f ) were treated with doxycycline from conception and harvested at two and three weeks of age . The doxycycline-treated tritransgenic mice were normal in birthweight and showed no gross extrapulmonary phenotypic or histologic abnormalities observed at 6 months . We examined activated c-Met ( phosphorylated c-Met ) by immunohistochemical staining in the airspace of doxycycline treated SPCMetf/f mice compared with bitransgenic or single transgenic controls . Alveolar epithelial staining for p-met was largely ablated in the SPCMetf/f mice , consistent with inducible compartmental deletion of Met ( Figure 1B , 1C ) . Modest and marked increases in airspace caliber were seen in two and three week old doxy-treated SPCMetf/f mice , respectively , ( Figure 1D , 1E ) . This finding suggested that c-Met expression in alveolar epithelial cells contributes to normal alveolarization . Since HGF is a known epithelial mitogen and survival factor , we investigated whether AECII abundance was altered in the SPCMetf/f lung . By immunohistochemistry , we found reduced number of AECII cells in the 1 month old SPCMetf/f lung ( Figure 1F and Figure S2 ) . As alveolar epithelial and endothelial morphogenesis are often interdependent , we examined microvascular abundance in the mutant lung . By thrombomodulin staining and quantitative immunohistochemistry , we found a marked truncation in the pulmonary vascular bed of the mutant mice ( Figure 1G , 1H ) . These data suggest that the architectural defects observed in the mutant mice are likely secondary to both the primary impairment in alveolar epithelial cell survival and secondary cell-nonautonomous effects on the microvascular bed . To further parse the alveolar epithelial cell phenotype of Met-deficient mice , we assessed measures of cell survival and stress in the alveolar compartment . The distribution of airspace proliferation , as assessed by Ki67 immunostaining , in the SPCMetf/f mice compared to wild type controls was different ( Figure 2A ) . Whereas Ki67 staining was predominantly localized to airspace epithelial cells in the wild-type lung , staining was most prominent in the alveolar macrophages of the mutant lung ( Figure 2A , 2B ) . By contrast , TUNEL staining , reflecting parenchymal cell death , was not enhanced in the mutant lung ( data not shown ) . Of note , this combination of reduced target cell proliferation without enhanced cell death was also observed in hepatocyte-specific Met deletion [13] . Because oxidative stress in the airspace compartment can reduce proliferation and increase airspace dimension , we examined nitrotyrosine staining in the lungs of SPCMetf/f mice compared to wild type controls . We found a more than 50% increase in nitrotyrosine staining in the mutant lungs ( Figure 2C , 2D ) . We assessed the expression of a panel of antioxidants in the lungs of mutant mice and found no significant change in the levels of NAD ( P ) H: quinone oxidoreductase-1 ( Nqo1 ) , heme oxygenase 1 ( Hmox1 ) and glutamate-cysteine ligase catalytic subunit ( Gclc ) but a trend towards reduction in mutant mice ( Revised Figure S3 ) . Oxidant injury often associates with inflammation , especially macrophage influx , in various models of parenchymal lung disease . We indeed found increased macrophage abundance and proliferation in the lungs of SPCMetf/f mice ( Figure 2E , 2F ) . Thus , the loss of c-Met expression in lung epithelial cells culminates in enhanced oxidative stress , reduced epithelial cell proliferation , mononuclear inflammation in the airspace compartment and a truncated microvascular bed . These insults likely confer the reduced AECII abundance and increased airspace dimension which define the airspace phenotype . Loss of alveolar c-Met expression does not affect extracellular matrix expression or abundance . Since airspace homeostasis requires extracellular matrix integrity and HGF is known to attenuate fibrosis in animal models , we assessed the deposition of elastin and collagen in the lungs of wild-type and Met-deficient mice . Trichrome and modified Hart's staining showed no altered deposition of collagen or elastin respectively in the lungs of mutant mice compared with age-matched controls ( Figure S4A , S4B ) . To establish whether HGF exerts a direct effect on epithelial activities involved in alveolar septation , we employed pharmacologic and genetic enhancement of HGF signaling in MLE12 cells , an established murine alveolar epithelial cell line . Treatment of cells with recombinant HGF induced proliferation which was more robust at higher concentrations than that seen with EGF treatment , a known epithelial mitogen ( Figure 3A ) . Similarly , transient transfection of human MET into MLE12 cells induced a significant increase in cell proliferation at 24 h when compared with vector control ( Figure 3B ) . Because epithelial cell scattering approximates the cellular migratory activity that is required for alveolar formation and can be mediated selectively by HGF [14] , [15] , we determined whether HGF could promote such behavior in MLE12 cells . Treatment of serum-starved MLE12 with recombinant HGF resulted in marked dispersion compared with media or EGF controls ( Figure 3C ) . Of note , the total cell counts were comparable in the EGF and HGF treated cells . Hepatocyte growth factor treatment also attenuated staurosporine-induced apoptosis in MLE12 cells ( Figure 3D , 3E ) . In order to determine whether HGF enhances cell survival of primary alveolar epithelial cells , we performed a survival analysis of isolated murine alveolar type 2 ( ATII ) cells from wild-type and SPCMetf/f mice . Others have shown that HGF induces proliferation of primary rat alveolar type II cells [16] , [17] . We found a significant increase in the survival of wild-type AEC cells at 48 h , consistent with a prosurvival effect of HGF signaling ( Figure 3F ) . Using primary cells , we examined whether antioxidant and antiapoptotic signaling contributed to the short-term prosurvival effect of HGF . We found a significant induction of the antioxidants Nqo1 and Gclc but no change in the expression of the antiapoptotic genes Bcl2 and Bax with HGF treatment ( Figure 3G and data not shown ) . This battery of in vivo and whole cell studies suggested that HGF/c-Met signaling utilizes both antioxidant and antiapoptotic effects to mediate selected components of the complex series of cellular events that are needed for alveolar formation and alveolar epithelial cell survival . Given the multiple alveolar epithelial responses conferred by HGF signaling , we queried whether augmented HGF signaling might induce a protective or reparative response in murine models of emphysema . TSK/+ mice , a spontaneously mutant strain heterozygous for a mutant allele of the matrix protein fibrillin-1 which compromises fibrillin-1 activity , are a well-accepted model of genetic emphysema . They display alveolar septation defects that evolve into overt emphysema [18] . We recently showed that the TSK/+ airspace phenotype is partially attributable to matrix-associated susceptibility to oxidative stress resulting in alveolar cell death [19] . Before proceeding with an HGF augmentation strategy , we determined whether there are alterations in HGF expression and signaling in the TSK/+ mouse model of impaired septation . Real-time PCR , ELISA analysis and immunoblotting of whole lung specimens showed no alteration of HGF and c-Met expression in the PD14 TSK/+ lung compared with age-matched controls ( Table S1 , Figure S5A , S5B ) . However , at 2 months of age , activated HGF levels were reduced in the TSK/+ lung . By immunohistochemical analysis , although we saw some regions of reduced c-Met expression there was overall no consistent reduction in c-Met expression in the PD14 TSK/+ lung ( Figure S5C , top ) . By contrast , we observed reduced and discontinuous expression of HGF within the interstitium of the airspace compartment of TSK/+ mice ( Figure S5D , bottom ) . Given the antifibrotic effects of HGF in rodent models and the proposed role of alveolar myofibroblasts in alveolar homeostasis , we used alpha smooth muscle actin immunohistochemistry to gauge the abundance of alveolar myofibroblasts in the TSK/+ lung . We found few myofibroblasts in the alveolar compartment of both wild-type and TSK/+ mice as well as preserved abundance of SMCs in the airway submucosa ( Figure S5E ) . These data suggest that airspace disorders characterized by abnormal extracellular matrix composition ( e . g . TSK/+ mice ) may exhibit altered HGF activation and deposition and that the TSK/+ mouse is an excellent model system to examine the therapeutic effects of HGF augmentation . Since alveolar cell specific deletion of Met compromises alveolar formation , we examined whether enhanced HGF signaling might rescue the airspace phenotype in TSK/+ mice . A subcutaneous pump containing active , recombinant human HGF , kindly provided by Drs . Ralph Schwall and Mark Merchant at Genentech , or carrier protein was inserted into adult TSK/+ mice and wild type controls . The HGF pumps delivered 50 µg/day over a two week period . We measured human HGF levels in HGF pump mice by ELISA assay , comparing intratracheal and subcutaneous pump delivery of comparable doses ( 50 µg/d for 3 d ) . A marked elevation in serum HGF was observed after pump delivery but not intratracheal delivery ( Figure 4A ) . Immunohistochemical staining for enhanced HGF signaling in the lung as evidenced by activated c-Met ( p-Met ) expression showed increased staining in the airspace compartment in the HGF-treated mice ( Figure 4B ) . We assessed airspace morphology as an index of airspace protection and repair in these mice . The TSK/+ mice treated with short-term HGF ( low dose and high dose ) demonstrated >17% improvement in airspace caliber ( Figure 4C , 4D ) . We also found reduced alveolar oxidative stress by both nitrotyrosine in the HGF-treated mice suggesting that HGF is able to antagonize oxidative stress ( Figure 4E , 4F ) . Using immunoblotting , we analyzed downstream signaling patterns that corresponded to the morphologic and oxidative stress rescue in the TSK/+ mice after HGF augmentation ( Figure 4G ) . We found increased stat3 and akt activation in the TSK/+ lung after HGF augmentation . Since activation of akt and stat3 associated with reparative effects of HGF on pathologic airspace enlargement , we assessed phosphoprotein activation in MLE12 cells treated with HGF . Treatment of MLE12 cells with recombinant human HGF induced activation of ERK , JNK and akt ( Figure 5A ) . Notably , we saw no stat3 activation with HGF treatment ( Figure S6A ) . We assessed whether akt was involved in critical prosurvival events by using staurosporine treatment of MLE12 cells to induce apoptosis . Wortmannin , a known akt inhibitor , inhibited HGF induced akt activation but not ERK activation in MLE12 cells ( Figure S6B ) . We found that staurosporine-induced apoptosis was inhibited by HGF treatment but pretreatment with an akt inhibitor wortmannin fully blocked the protective HGF effect ( Figure 5B ) . Taken together , the in vivo and in vitro findings not only suggest an important role for HGF/c-Met signaling in alveolar epithelial cell survival and maintenance of postnatal airspace homeostasis but also a role for prosurvival signaling cascades as mediators of these events .
Our investigation of the loss of function phenotype attached to the cell-specific deletion of Met identifies its activation in AECII cells as an important cell-autonomous event in alveolar formation . How does HGF promote alveolarization ? The major consequence of Met deletion in airspace epithelial cells is the reduced abundance of these cells in the juvenile and adult lungs , increased oxidative stress and inflammation and truncation of the vascular bed . Since whole cell data by our lab and others suggests that c-Met activation induces cell survival and enhanced migration in primary and immortalized alveolar epithelial cells , the septation defect is likely attributable to these mechanisms that converge to compromise septation [8] , [27] . Recent work by Factor et al examining mice with hepatocyte- and liver-specific deletion of Met similarly demonstrated a profound cell autonomous defect in cell cycle progression , invoking an Erk-1 dependent mechanism [12] . An additional aspect revealed by our study is the truncation of the pulmonary microvasculature which accompanies loss of epithelial expression of c-Met . Yamamoto showed that mice deficient in lung epithelial VEGF-A displayed impaired pulmonary capillary formation and reduced HGF production , implicating the c-Met pathway as a critical mediator of epithelial-endothelial crosstalk in lung homeostasis [10] . Similarly , our findings of combined therapeutic and developmental effects of HGF in the lung support a critical homeostatic role for HGF signaling in the airspace that likely incorporates proliferative , migratory and morphogenic agendas and distinct prosurvival pathways . In the lung , hepatocyte growth factor , secreted by endothelial cells , epithelial cells and interstitial fibroblasts , is sequestered in a precursor state in the extracellular matrix . Although the precise activation events are not well understood in the lung , inflammatory insults trigger the liberation of active HGF and the initiation of HGF/c-Met signaling in many tissues [28] , [29] . If HGF signaling is a homeostatic mechanism needed for the maintenance of airspace morphology and cell composition , then an effective deficit in HGF signaling may occur in airspace disorders that are not marked by inflammation . We show that active HGF levels are reduced but c-Met protein levels are preserved in the TSK/+ lung compared with wild-type controls . We suspect this is secondary to defective HGF deposition and activation on the abnormal TSK/+/+ extracellular matrix evident by HGF immunostaining ( Figure S2C ) . Accordingly , downstream HGF signaling is impaired in TSK/+ mice . Since TSK/+ mice exhibit marked postnatal oxidative stress in the lung that promotes airspace enlargement , this lack of maintained or enhanced HGF signaling may be a cause of the enhanced oxidant stress and lead directly to the airspace phenotype [19] . The improved airspace caliber resulting from HGF administration in the TSK/+ mouse suggests suboptimal HGF signaling may be a hospitable context for for airspace protection/repair with HGF administration . As stated above , reduced p-Met activation in the TSK/+ lung combined with enhanced apoptosis and oxidative stress is consistent with impairment in both proximal HGF/c-Met signaling and reparative downstream pathways . Investigators report both reduced and maintained HGF levels in the lungs of patients with COPD/emphysema [30] , [31] . Interestingly , patients with acute lung injury typically have increased HGF levels in the bronchoalveolar fluid reflecting a reparative response [32] . Whether those levels are maintained as theinjury evolves is unknown . Infants with bronchopulmonary dysplasia who have reduced HGF levels typically have worse outcomes [33] . Thus , the lack of an increase in active HGF signaling in the TSK/+ lung likely reflects an impaired response to epithelial injury . Evidence that selective cytokines which contribute to alveolar morphogenesis , such as epidermal growth factor ( EGF ) , fibroblast growth factor 10 ( FGF10 ) , platelet derived growth factor A ( PDGFA ) , and vascular endothelial growth factor ( VEGF ) , have protective or therapeutic efficacy for animal models of adult airspace disorders is limited [10] , [34]–[36] . This limitation largely reflects the extraepithelial effects of these cytokines that may antagonize normal lung repair ( reviewed in [37] ) , a therapeutic requirement for the neonatal rather than adult milieu or simply absence of well-constructed preclinical trials . For example , VEGF is required for airspace formation but overexpression or pharmacologic augmentation of VEGF can have injurious effects , nicely discussed in [38] . Although without a clear role in alveolar formation , keratinocyte growth factor ( KGF ) is the only growth factor which has a similar functional repertoire as HGF . However , a major limitation for the therapeutic potential of KGF is the probable requirement for pre- or concurrent injury administration ( protective rather than therapeutic effects ) [3] , [39] . In fact , since overexpression of KGF in the murine lung results in severe malformations , the protective/therapeutic window must be carefully defined in neonatal or developing mice [40] . We show that HGF administration has reparative effects in adult TSK/+ mice suggesting a more flexible therapeutic repertoire than KGF . We plan to dissect this difference in future studies . Growth factor induced repair of epithelium may incorporate proliferative , antiapoptotic , migratory and morphogenic agendas [5] , [41] . These converge to alter cellular turnover and increase cellular survival , permitting the regeneration of functional structures . We found in the TSK/+ model that stat3 and akt activation , known prosurvival mediators , correlate with the maintenance or reestablishment of airspace integrity . HGF/c-Met signaling induces stat3 activation frequently resulting in both cellular migration and morphogenesis in a variety of cell systems [42] . Further , a loss of stat3 activation in the murine lung increases susceptibility to hyperoxic injury and overexpression of an activated stat3 confers protection [43] , [44] . Similarly , akt is involved in airspace maintenance in a neonatal model of lung injury [45] , [46] . HGF-induced akt activation ameliorates cigarette smoke extract induced epithelial cell death [47] . We show here that HGF treatment of immortalized murine lung alveolar epithelial cells ( MLE12 ) , an established model of AECII cells , activates akt and appears to mediate prosurvival signaling . We also show that HGF treatment of primary alveolar type II cells promotes survival and expression of antioxidants . Future efforts will focus on dissecting the interface between prosurvival signaling and antioxidant protection in the airspace compartment . In summary , although alveolar septal loss is the most intractable functional and anatomic lesion in COPD , the molecular basis of this process remains elusive . The mitogenic , motogenic and morphogenic features of HGF make it an attractive candidate mediator of airspace repair . We propose that reduced c-Met activation and expression underlie the inadequate reparative response in the emphysematous lung . Mice deficient in Met expression in alveolar epithelial cells display compromised epithelial cell abundance , pruning of the microvascular bed and airspace enlargement . Reduced HGF expression and c-Met activation are evident in inbred mice with genetic emphysema . We have also found that pharmacologic augmentation with recombinant HGF in a murine model of emphysema results in both reduced oxidative stress in the airspace and improved airspace dimension . We define here an important homeostatic role of HGF signaling in airspace formation , maintenance and regeneration suggesting that the HGF/c-Met pathway should be explored for airspace disorders such as bronchopulmonary dysplasia and emphysema .
Adult C57Bl6 and TSK/+ mice were housed in a controlled environment and provided with standard water and chow . Animal care was in compliance with IACUC recommendations . Mice conditionally deficient in Met expression , SPC-rtta/+;otet-Cre/+;metf/f , in alveolar epithelial cells were generated by crossing Metf/f mice harboring an inactivating conditional deletion of exon 16 of the mouse Met gene [12] with bitransgenic mice expressing SPC-rtta;otet-Cre [11] . Pups resulting from these matings were produced in comparable litter sizes and the genotypes represented in Mendelian ratios . Controls were bitransgenic or single transgenic mice . The mice were housed in a facility accredited by the American Association of Laboratory Animal Care , and the animal studies were reviewed and approved by the institutional animal care and use committee of Johns Hopkins School of Medicine . To induce Cre recombinase , mice were treated with doxycycline ( Sigma ) at 5 mg/ml in drinking water from conception to time of harvest . Tight-skin ( TSK/+ ) mice backcrossed into a C57Bl/6 background without the pallid allele were generated and maintained as described [19] . Mice were genotyped using standard protocols [11] , [12] , [48] . Recombinant HGF provided by Genentech was administered through an intraperitoneal miniosmotic pump placed under isoflurane anesthesia . The pumps were loaded with HGF with carrier in PBS or carrier in PBS alone ( control ) producing a total daily amount of 25–50 µg for 2 weeks . For intratracheal delivery , HGF was administered per catheter inserted into the tracheal under isoflurane anesthesia and direct inspection . Results are expressed as means ± SEM unless otherwise stated . Comparisons between 2 experimental groups were examined using the Student T test or Mann-Whitney rank sum test . Comparisons among 3 or more groups were performed by one-way ANOVA . All statistical analyses were performed with Sigmastat ( version 3 . 5; systat Software , Chicago , IL ) . A p<0 . 05 was considered significant . Additional and more detailed methods are provided in Text S1 .
|
The airspace compartment of the mammalian lung , comprised of spherical sacs termed alveoli , harbors the architecture , cellular composition , and molecular armamentarium to perform the critical function of gas exchange or oxygen uptake . Despite the necessity of this alveolar compartment for organismal viability , the mechanism by which alveoli are formed and maintained is obscure . Furthermore , no treatments are currently available that can regenerate the airspace once damaged . In this manuscript , we sought to determine whether hepatocyte growth factor , a cytokine with a functional armamentarium that subserves the critical events of alveolar formation ( epithelial proliferation , migration , resistance from apoptosis and angiogenesis ) , could be an important mediator of alveolar formation and airspace maintenance . Our simple paradigm was that critical homeostatic pathways for the lung should operate both in lung formation and in lung maintenance/regeneration . Using an informative battery of mouse models and cell lines , we show that hepatocyte growth factor is a determinant of alveolar formation and that the enhancement of hepatocyte growth factor signaling can both protect and repair the airspace from pathologic airspace enlargement or emphysema .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"anatomy",
"and",
"physiology",
"pulmonology",
"animal",
"models",
"developmental",
"biology",
"histology",
"model",
"organisms",
"molecular",
"genetics",
"morphogenesis",
"respiratory",
"medicine",
"biology",
"regeneration",
"physiology",
"genetics",
"genetics",
"and",
"genomics"
] |
2013
|
Hepatocyte Growth Factor, a Determinant of Airspace Homeostasis in the Murine Lung
|
Neural coding through inhibitory projection pathways remains poorly understood . We analyze the transmission properties of the Purkinje cell ( PC ) to cerebellar nucleus ( CN ) pathway in a modeling study using a data set recorded in awake mice containing respiratory rate modulation . We find that inhibitory transmission from tonically active PCs can transmit a behavioral rate code with high fidelity . We parameterized the required population code in PC activity and determined that 20% of PC inputs to a full compartmental CN neuron model need to be rate-comodulated for transmission of a rate code . Rate covariance in PC inputs also accounts for the high coefficient of variation in CN spike trains , while the balance between excitation and inhibition determines spike rate and local spike train variability . Overall , our modeling study can fully account for observed spike train properties of cerebellar output in awake mice , and strongly supports rate coding in the cerebellum .
Transmission of information through firing rate changes in populations of connected neurons is one of the most widely accepted principles of neural coding . In motor control , for example , cortical neurons showing firing rate changes as a function of movement direction can be said to dynamically compute the current movement direction in a population vector [1] . This representation also works well computationally in abstract neural networks , for example when the motion of handwriting control is computed in the neural engineering framework [2] . Little is known , however , about how biological neurons utilize rate codes transmitted by their typically hundreds or thousands of input synapses to control their own output firing rate , and how robust such a code is in the presence of noise , intrinsic non-linearities given by voltage-gated channels , and a balance of excitatory and inhibitory inputs . Further , it is unclear whether rate codes are equally present in inhibitory as in excitatory transmission . We addressed these questions by studying the inhibitory transmission between cerebellar cortical Purkinje cells ( PCs ) and their targets in the cerebellar nuclei ( CN ) through recordings from awake mice and detailed biophysical simulations of synaptic integration in CN neurons . Linear rate coding has been identified to represent excitatory input information from granule cell input in PCs [3 , 4] , but the correlation of coding at the population level and its transmission to CN neurons in vivo remains unclear . We use rhythmic motor patterns and in particular the rhythmic control of respiration as a model behavior to study the transmission of rate coding in cerebellar circuits , as rhythmic respiratory rate modulation is well expressed in the spiking activity of PCs in the cerebellar vermis [5] as well as in the synaptically connected medial ( fastigial ) cerebellar nucleus [6] , and this pathway plays a functional role in the neural control of respiration [6] . In the present study we used an updated version of a full biophysical model of CN neurons [7] to study how the population of Purkinje cell inputs expected to converge on a single CN [8] may transmit a rate coded rhythmic behavior , and whether CN model generated spike trains can account for spiking properties recorded from CN neurons in awake mice . We developed a new algorithm that allows the flexible construction of sets of artificial PC spike trains that match the statistical properties of recorded PCs while also allowing the insertion of correlations observed between pairs of recorded PCs into a larger set of PC spike trains that converge onto a single CN neuron as input . This new algorithm development was necessary because it is at this time physiologically impossible to record from and identify all the PCs that converge onto a single CN neuron . Therefore , in order to simulate a realistic range of rate-correlations and respiratory coding correlations between the ~50 PC inputs received by a single CN neuron , it is necessary to generate populations of artificial spike trains ( ASTs ) in which each AST matches the statistics of PC recordings ( which we obtained from awake mice ) while flexibly allowing the addition of specific rate co-variances between ASTs . We achieved this goal by creating an intermediate representation of spike trains as rate templates that could be manipulated algebraically to show more or less rate co-variances both for respiratory related rate changes and slow rate fluctuations . To create ASTs we could then draw gamma distributed interspike intervals from the rate template to match the template’s rate fluctuations as well as the recorded spike train statistics . To our knowledge this study presents the first such algorithm , which we expect will be generally useful for similarly minded modeling studies of synaptic integration in the awake brain . Our CN modeling results for the first time give a full match of CN spiking properties seen in awake recordings derived from the biophysical properties of CN neurons and the statistics of their synaptic inputs . The results reveal an unexpected amplification of rate coding at the CN output compared to the PC inputs received and show a highly robust transmission of rate codes from the cerebellar cortex to the CN via inhibition in the waking condition . They also provide evidence for an involvement of intrinsic cellular dynamics in providing gain control in the transmission of rate codes .
The starting point of our analysis was a database of 21 PC , 11 mossy fiber ( MF ) and 16 CN recordings . These data were obtained in awake head-fixed mice with multiwire recordings while respiration was monitored using a thermistor placed in front of one nostril [5 , 9] . Out of 20 PCs that were analyzed for rate modulation linked to respiration , 15 ( 75% ) showed significant rhythmic rate modulation , as indicated by a deflection of the rate change in a peri-event time histogram ( PSTH ) triggered on inspiratory event markers above 3 standard deviations ( S2 and S3 Figs ) . Standard deviations were calculated from a set of 100 control PSTHs from each cell that were calculated from randomly shifted spike time series with respect to the respiratory event markers . The same analysis showed significant respiratory modulation for 10 of 16 CN neurons ( 63% ) , and 6 of 9 ( 67% ) MF recordings . This strong representation of respiratory activity supports previous evidence that the vermal cerebellar cortex through its output connection in the medial cerebellar nucleus is involved in the adaptive control of respiration [6] . The rate modulation for different cells showed different phase relationships to respiration , and the averaged rate modulation in the PC , CN , and MF neuron population was not significant ( S3 Fig ) , suggesting that cerebellar respiratory modulation occurs at all phases of respiration to a similar degree , though in different populations of neurons . We also made a detailed analysis of the recorded PC , MF and CN baseline spike train statistics , in particular firing rate as a function of time , interspike-interval ( ISI ) distribution , coefficient of variation ( CV ) , and local variation ( LV ) [10] , which indicates the variability of pairs of successive ISIs ( S1 Fig , Table 1 in S1 Text ) . Our goal was to determine whether the spike train statistics and respiratory modulation of CN neurons can be explained from the dynamics of a biophysically realistic CN neuron model [7] and the input patterns received . To achieve our goal we first had had to design a bootstrapping method by which to extrapolate from 2 simultaneously recorded PCs to a population of ~50 PC spike trains with flexible rate covariances that converge on a single CN neuron with strong synapes [8] . We determined that our recorded PC spike trains had broad cross-correlations that were in part related to behavior [9] , but did not find any millisecond precision in simple spike cross-correlations here or in previous studies [11 , 12] . Complex spikes were removed from the PC spike trains , and not further considered in this study . We constructed a Matlab ( MathWorks , Inc . ) algorithm by which we can assemble artificial spike trains ( AST ) closely matching properties of single recorded PCs ( Fig 1 ) . The core of this algorithm consists of building and manipulating spike rate templates ( Fig 1A and 1B ) , which are constructed by convolving spikes recorded from a single neuron with Gaussians [13 , 14] ( see Methods ) . To construct ASTs we draw gamma distributed ISIs from a distribution with a mean rate tracking a rate template , and a shape parameter kappa ( κ ) that is mathematically derived from the LV of the recorded spike train , where for gamma distributed events LV = 3 / ( 2 κ +1 ) [10] . To obtain an ISI distribution in an AST that matches the original recording ( Fig 1C ) we further had to perform a refractory period correction , as gamma distributions do not model processes with refractory periods directly ( see Supplemental Methods for details ) . We validated our ASTs by comparing the spike train power spectrum between recorded neurons and the built-to-match ASTs ( Fig 1D ) , and by ascertaining that the coefficient of variation ( CV ) and the LV of the AST also matched the recording closely ( Fig 1E ) . An important observation was that the LV could be modeled as a static parameter as previously observed for cortical neurons [15] , but the global variability of the spike train over time represented by the CV is an outcome measure that is influenced by the LV as well as the spike rate modulation over time . Using these methods we made populations of 50 PC ASTs with statistical properties and spike rate fluctuations matching our recorded PCs while also being able to flexibly control rate covariances . All 50 PC ASTs used as input to the CN model were taken from the same master rate template of a single PC with specific different manipulations of rate-covariances for different simulation runs as described below ( also see Supplemental Methods for details ) . These AST populations were then used to analyze how convergent input from 50 PCs would influence CN spiking , and what properties of convergent input were needed to account for observed CN spike train statistics . The biophysical CN model we used consists of 485 dendritic and one somatic compartments incorporating 9 active conductances to replicate slice CN recordings [7] . We included the modifications of ion channel voltage-dependence and density as well as synaptic kinetics described in the supplemental materials of the original publication ( [7] , S3 & S4 Figs ) , which lead to a more depolarized level of tonic depolarization ( Fig 2A ) and a more linear f-I curve ( Fig 2B ) as well as faster synaptic kinetics to more closely replicate CN slice recordings in these qualities [16 , 17 , 18 , 19] . In the present study we further modified the synaptic kinetics of PC->CN synapses to incorporate the experimentally determined short term depression parameters [20 , 21] leading to a steady state depression of around 60% for a Purkinje cell firing rate of 75 Hz ( Fig 2C ) . The resulting spiking pattern with random excitatory and inhibitory input trains of the modified model remain similar to the original publication ( Fig 2D and 2E ) , and are based on a balance of excitatory and inhibitory input currents with a fluctuating total synaptic current near zero ( Fig 2F ) , which modulates the spontaneous activity of these neurons [16 , 22] . Next we characterized the CN model spiking output statistics for input patterns aimed to match the PC spike train statistics derived from our recorded data . We used 50 PC ASTs ( 48 dendritic , 2 on the soma ) to match the number of strong PC inputs to converge on a CN neuron recently described [8] . We also applied 48 dendritic mossy fiber ASTs to create the required balance between excitation and inhibition [16 , 22] . We scanned through an array of input parameter settings that are not fully experimentally constrained , notably the size of unitary excitatory and inhibitory conductances ( Gin and Gex ) , and the amount of rate covariances present between 50 synchronous PC inputs . The latter setting was manipulated by a shift fraction ( SF ) , that is the proportion of rate modulation that utilized a randomly time shifted version of the master rate template . For the first set of simulations we used the PC firing rate of the template neuron ( 64 . 9 Hz ) for all 50 ASTs . MF inputs to the model were also taken from a typical single recorded MF rate template , but as this study focused on the effect of rate covariances present in the PC pathway to influence CN spiking statistics we chose to use a SF of 1 . 0 for our baseline simulation ( all MF inputs are temporally decorrelated ) and an MF input rate of 20 . 4 Hz , which is the recorded sample mean . The results of this input parameter scan show that using different ratios of Gin and Gex allowed us to achieve a wide range of CN output firing rates ( Fig 3A ) , and revealed a systematic relationship between firing rates , CV and LV ( Fig 3B and 3C ) such that faster CN spike trains associated with a smaller Gin / Gex ratio showed a lower CV and LV despite using the same PC input spike trains . Further , the CV and LV of CN spike trains were higher for larger absolute values of Gin ( Fig 3B and 3C , red traces ) . For a high Gin ( 20 nS per PC input ) the PC input rate covariance also had a strong effect on the CN output CV , such that a higher input rate covariance ( Fig 3B , red traces with x symbols ) resulted in a higher CN spike train CV . In contrast , the LV of CN spiking was much less affected by the input rate covariances ( Fig 3C ) . A key result of our study is given by the match of the dependencies between CV and LV of CN spike trains between our simulations ( Fig 4A and 4C ) and our recorded CN data sample ( Fig 4B and 4D ) for the full range of physiological spike rates between 10 and 70 Hz . This simulation result indicates that the statistics of the PC and MF input to the CN as derived from our PC and MF recordings can fully account for the CN spike train statistics recorded in the same state . Interestingly , the match between recordings and simulation was best for a simulated SF of 0 . 5 , indicating that the spike train statistics in the CN recordings are most compatible with PC input that contains about 50% rate covariance . Further , the variability between our CN recordings can be explained by a possible variability in total PC input conductance amplitudes received by different CN neurons and different rate covariances between these inputs . These results for the first time fully account for spike train statistics in the awake state in a biophysically based neural simulation . While certainly other factors than the PC input statistics can influence CN spike statistics in the animal , our results demonstrate that the PC input statistics alone are sufficient to account for the full spectrum of recorded CN rates and LV as well as CV statistics and their interdependence . Next , we aimed to incorporate the respiration related rhythmic spike rate modulation in the PC input to CN neurons in our simulations to determine whether the recorded PC respiratory modulation ( Fig 5A and 5B ) can explain the recorded CN modulation ( Fig 5E and 5F ) . An important question not addressed by our recordings concerns the required level of covariance in respiratory modulation between PC inputs to a single CN to allow for the observed amplitude of CN respiratory modulation if it was solely transmitted by PC inputs . In order to create ASTs with respiratory modulation matching the recordings we again employed rate template manipulations . We determined the average rate modulation triggered by respiration in a given PC ( Figs 5A , 5B , S2 and S3 ) , and then we convolved the normalized rate modulation waveform with our master rate template at the measured time of each inspiration . We find that by drawing random gamma spike trains with refractory periods from this combined rate template we are able to create PC ASTs with respiratory rate modulation closely matching the experimental data ( Fig 5C and 5D ) while maintaining the spike train statistics of recorded PCs including their rate , LV , CV and power spectrum . As a proof of concept simulation we picked a specific CN recording with a peak of 36% spike rate increase during respiration ( Fig 5E and 5F ) and for our simulation input picked a Gin of 16 nS and Gex of 3 . 5 nS , which we knew from our parameter scan to result in a matching mean baseline CN simulation spike rate of ~22 Hz . We then asked the question of how many of the 50 PC inputs need to show the respiratory modulation shown in our ‘typical’ PC recording ( Figs 5A , 5B , S2A–S2C and S3A–S3C ) in order to generate the behavioral modulation strength seen in our ‘typical’ CN recording ( Figs 5E , 5F , S2D–S2F and S3D–S3F ) . The results showed that a respiratory modulation in 25 PC inputs ( i . e . 50% of inputs ) resulted in a match with our recorded CN modulation ( Fig 5E–5H ) . Next , we determined the robustness and relative expression strength in the transmission of respiratory rate modulation in the PC -> CN pathway by systematically varying the number of modulated PC inputs and the strength of modulation in each input for a slow and a fast spiking CN simulation resulting from 2 different levels of excitation ( Fig 6 ) . We find that a change in the PC respiratory modulation strength is transmitted faithfully to the CN , and that respiratory modulation is well transmitted by slow or fast firing CN neurons ( Fig 6A ) . Both the fraction of modulated PC inputs ( BMF ) and the strength of PC respiratory modulation ( BMS ) had strong effects on CN modulation ( Fig 6B ) . At the strength of PC modulation present in our experimental sample PC ( Fig 5A ) , a modulation of 10 of 50 PC inputs ( BMF = 0 . 2 ) to the CN neuron was sufficient to result in a significant output modulation . If all PC inputs to the CN simulation were modulated using the 11 . 4% mean rate decrease in the PSTH trough observed in the sample PC , the CN mean PSTH peak rate increase was 25 . 9% at a firing frequency of 60 Hz , and 48 . 3% at a firing rate of 20 Hz , indicating that the respiratory modulation depth is amplified in the transmission from the PC to the CN in an inhibitory transmission . Strong respiratory modulation in the CN lead to a moderate increase in the CV of the CN spike trains ( Fig 6C ) , while the LV was less affected ( Fig 6D , Gex = L . red solid lines ) . We further examined the effect of global rate covariances between PC inputs on respiratory modulation ( SF 1 . 0 vs . 0 . 5 ) , and found that this manipulation of background rate covariance only had a small effect on the transmission of respiratory modulation ( Fig 6B , circle vs asterisk symbols ) , while it had a strong effect on the overall CV of the spike train ( Fig 6C ) . As detailed in the Supplemental Information we found that the transmission of respiratory modulation was also robust against different PC input firing rates , the presence of absence of short term depression in the PC-> CN synapses , using rate templates from a different PC , and changing the gain on template rate fluctuations ( S2–S7 Figs ) . The key outcome of these sets of simulations was that the inhibitory PC inputs on tonically active CN neurons provide a sensitive and accurate means of transmitting a rate code related to controlling behavior , and that the strength of this rate transmission is highly dependent on the fraction of inputs modulated with the same time course . Further , the transmission of rate modulation is robust in the face of common background rate modulation in the input and operates well for the full range of observed CN firing rates ( 10–70 Hz ) . Importantly our simulations demonstrate that the observed respiratory modulation in CN neurons can be fully explained by the measured rate modulation in PCs if at least 20% convergence of similarly modulated PCs onto single CNs is present . Earlier anatomical estimates of the number of PC inputs on CN neurons [23] were much higher than given by the recent physiological assessment [8] . While the recent work can account for the larger number of boutons anatomically observed by positing multiple boutons per PC input to a CN neuron , we were interested to know what the consequence of using 500 instead of 50 inputs would be for matching our recorded CN data from awake mice . We created 500 ASTs using the same rate template as previously for 50 , but we divided the unitary synaptic conductance by 10 to arrive at a similar average conductance waveform ( Fig 7A ) . A notable difference in the total conductance of 500 inputs was that high frequency fluctuations were much diminished due to averaging over 500 instead of 50 random processes . Notably , this had a large effect on the output spike rate from the CN simulation ( Fig 7B and 7C ) , which was diminished for 500 inputs from 63 Hz to 20 Hz for a high level of excitation , and from 20 Hz to near zero for a low level of excitation . This dramatic difference illustrates the high importance for fast input conductance fluctuations in triggering individual sodium action potentials , a property not seen in integrate and fire neurons . We have previously also observed this finding in dynamic clamp experiments of CN neurons in brain slices [16] . Despite the large decrease in CN spike rate for 500 PC inputs , the respiratory modulation remained strong ( Fig 7D–7F ) , and the absolute values of respiratory spike rate increases were nearly the same for a 20 Hz spike rate with 500 PC inputs than they were with a 60 Hz spike rate with 50 PC inputs with the same spike train properties ( Fig 7E and 7F Gex: high , dashed lines ) . These findings again show that inhibitory synaptic transmission is a highly robust carrier for a behavioral event related rate code . The main computational outcome of using 500 instead of 50 PC inputs was that much more excitatory input is needed in order to match the spike rates in the model with those recorded in awake mice . Finally , we asked the question whether the intrinsic active currents of CN neurons make an important contribution to the spiking statistics and respiratory modulation in our simulations . While a full treatment of this question falls outside the scope of this study , we used manipulations of the density of the calcium dependent potassium current ( SK ) to see what contributions this modulatory current makes to CN coding properties in awake mice . In previous work we and others have shown that this current is present in CN neurons and that blocking it with apamin causes bursting , bistability and pronounced spike rate increases with depolarization [18 , 24 , 25] . Further , stochastic excitatory and inhibitory input patterns in the CN model lead to strong fluctuations in SK current [26] . The involvement of this important modulatory current in synaptic integration in the awake animal remains unknown , however . Our default simulation made to match typical CN slice recordings from 14-21d old rats had a somatic SK density of 2 S / m2 and a dendritic density of 0 . 6 S / m2 . We varied these densities for SK densities between 0 and 8 S / m2 in the soma and proportionally 0 to 2 . 4 S / m2 in the dendrites ( Fig 8 ) . When SK was absent , comparing simulation Vm traces for 0 vs 8 nS SK density with the same synaptic input , we find a much reduced spike-afterhyperpolarization ( example indicated by blue arrow in Fig 8A ) and much stronger spike rate modulation for a given input rate modulation ( Fig 8A , see 0 . 7 to 0 . 9s for a period of decreased inhibitory conductance ) , as should be expected from the biophysical properties of this potassium current that is activated via the calcium inflow with each action potential . Not surprisingly , there is also a systematic decrease in overall spike rate with increasing SK density ( Fig 8B ) and a decrease in CV ( Fig 8C ) . The LV on the other hand shows a non-monotonic dependency on gSK , with a maximum near 4 nS ( Fig 8D ) . While SK is known to regularize spike trains [27] , this usually refers to the CV . The low LV when SK is absent is probably due to the high local regularity during periods of high frequency firing , but we did not further examine this effect . The effect of SK density on respiratory rate change transmission was also strong ( Fig 8E and 8F ) . With increasing gSK the CN output PSTH rate modulation with the same input was much diminished . This result indicates that SK is well suited to dampen the transmission of behaviorally related rate changes . Interestingly SK appears to be downregulated in adult rodents [25] , suggesting that as the cerebellum matures the gain of rate change transmission may be increased . Overall the strong effect of SK on spiking statistics and synaptic transmission of rate changes shows the high importance of intrinsic neuronal properties on the transmission of behaviorally related rate codes .
Our study posed the general question on how rate codes can be transmitted by inhibitory synaptic inputs using the cerebellar cortex to cerebellar nuclei projection as a paradigmatic example . This inhibitory connection is particularly interesting in that it conveys the entire output from the cerebellar cortex , and the cerebellar cortex is commonly thought to be involved in coding detailed temporal aspects of motor behavior [28 , 29 , 30] . Therefore , detailed temporal information has to be transmitted through the inhibitory cerebellar cortico-nuclear pathway . However , cerebellar research has generated conflicting ideas on whether this information is transmitted by a rate code [3 , 4] or by a temporal code triggered by input synchronicity [8 , 31] . While the distinction between rate and temporal codes can be blurry at intermediate values of temporal precision , one would generally take neural algorithms depending on coincidence detection [32] , synfire chains [33] or input synchrony to detect patterns [34] as examples of a temporal code , while spike rate modulation at the time scale of the behavior controlled ( which could be quite fast for saccades for example ) represents a rate code . Our findings with respect to the coding of respiration in mice in this study are fully supportive of the rate coding model in the control of cerebellar output , as rate was smoothly varying on the time scale of the behavior observed . Our findings substantiate the concept that inhibitory synaptic transmission can convey such information with high accuracy in tonically active neurons . Nevertheless , it is entirely possible that a temporal code is multiplexed with this rate code , and would be triggered by specific events , such as motor errors . In the cerebellum such an event in particular is likely to be coded by highly synchronous climbing fiber firing [3 , 4] , which could result in rebound activity in the CN [7 , 35 , 36 , 37] . This pathway should be analyzed carefully in future modeling work , but an experimental database of simultaneous cerebellar cortical and nuclei recordings in behaving animals while assessing climbing fiber synchrony is not yet available . For simple spike activity in cerebellar cortex in our mouse preparation we previously described an absence of synchronized spiking or synchronized pauses with respect to respiration and licking [5] , or sensory activation in anesthetized rats [12] . Our modeling results in the present study show that indeed such coincident PC simple spike inputs to a CN neuron are not required to explain the observed rate , regularity or respiratory modulation of our CN recordings . Instead , we found that rate coding of PCs is fully sufficient to account for observed CN spiking properties , but that a substantial correlation in the rate modulation between PCs projecting to the same CN neuron is required . We used a detailed biophysical CN neuron model to perform our investigation , which allows us to address the question of how much intrinsic active properties of CN neurons are important in decoding synaptic input . Interestingly , in the present input scenario of a time-varying balance of excitatory and inhibitory input the strong rebound firing capabilities of the model , which match experimental findings [7] , did not come into play as significant de-inactivation of the rebound currents ( T-type Calcium and persistent Na conductances ) through strong hyperpolarization did not occur . Nevertheless , our investigation of the role of the SK conductance in the present study shows that the neurons’ active properties are highly significant in decoding synaptic input . Specifically , the SK conductance in CN neurons is known to cause prolonged spike-afterhyperpolarizations and regularize spontaneous spiking [17 , 24] , which after block of SK current with apamin becomes highly bursty [17 , 24] . In a previous dynamic clamp study we showed that bursting is suppressed with a baseline of inhibitory and excitatory input conductance , but that the gain of responses to input modulation was increased when SK was low [25] . This role of SK controlling the gain of the synaptic response function was confirmed in our present modeling study for input conditions of respiratory spike rate modulation in the awake mouse . Recordings from slices of rodents at different ages suggest that SK is downregulated as animals become adult [25] , thus perhaps allowing a greater CN output modulation by input fluctuations as the cerebellum learns to code for specific behaviors . However , even in the adult mouse the amount of SK current observed in single CN neurons may be highly variable as is typically observed for voltage-gated currents [38] , and possibly serve as a gain control mechanism on the synaptic coding function of behavioral spike rate modulation that could be regulated through intrinsic plasticity . Such SK plasticity has not been studied in the CN , but is known to occur in other cell types [39] . While our results support the notion that modulated PC input on CN neurons is sufficient to explain observed CN spike train statistics and respiratory modulation , we do not wish to imply that mossy fiber inputs to the CN are irrelevant or ineffective in this regard . In our previous dynamic clamp studies in CN brain slice recordings we have shown that PC input alone can control CN spike rate and regularity in the presence of tonic excitation , which is required to achieve a necessary balance between excitation and inhibition [16] . However , when the MF input is also modulated in the dynamic clamp input to mimic in vivo input conductances [22] , the MF activity can also control CN spiking , and that MF and PC triggered modulation of CN spiking is roughly additive . Nevertheless , we found that due to the high proportion of slow NMDA conductance in MF input to CN neurons [40 , 41] that CN spike train irregularity is predominantly caused by PC input transients [22] . Our current results lead to the prediction that the contribution of MF input to respiratory rate modulation would critically depend on the amount of respiratory rate-covariance in the MF inputs to a CN neuron . MF respiratory modulation , similar to PC modulation , shows a variety of phase relationships to respiration ( S2 and S3 Figs ) , and therefore a mechanism to strengthen MF convergence with similar modulation on single CNs would be required . A detailed exploration of the required MF input parameters in order to be effective is outside of the scope of the present study , but will be undertaken in the future . Another implication that we do not wish to be taken from our modeling study is that the respiratory modulation transmitted from the PC to the CN does in fact control respiration . In fact , our working hypothesis is that baseline respiration is not controlled by the cerebellum , but that the observed coordination of different orofacial rhythms such as licking , swallowing , whisking with respiration [9] is effected through the connection of the medial cerebellar nucleus to the respective rhythm generators [42] . Establishing the functional role of cerebellar output on the coordination of these rhythm generators and the ability of the cerebellum to delay or advance the respiratory cycle when needed will require new experimental studies where these rhythms are challenged and the output of the cerebellum is optogenetically manipulated . Any given neuron in the brain typically receives synaptic input from hundreds of other neurons . For the synaptic transmission of a rate code it is therefore critically important to understand what number of these inputs need to be rate co-varying , in order for a robust transmission of behaviorally related information . This point is closely related to the question of how population coding is instantiated in the brain , as enough neurons need to be participating in the same coding process so that their convergent connections on a target population would transmit a rate code accurately . To our knowledge this study is the first that quantifies the answers to these questions in the framework of a biophysically accurate model to match a data set recorded in awake animals . We find that in the cerebellum where 50 PCs converge onto a single CN neuron , transmission of significant rate modulation required about ~10 ( 20% ) rate covarying PC inputs , while ~25 ( 50% ) PC inputs resulted in an outcome matching one of the stronger CN rate modulation amplitudes found in our experimental data . This code was found to be robust against interference from both correlated and uncorrelated background noise . These modeling results make a strong experimental prediction that populations of PC neurons converging to single CN neurons need to show a larger shared behavioral rate modulation than is present in a random sample of single recorded PCs . While such data are not yet available , advances in calcium imaging at single cell resolution in combination with transsynaptic retrograde labeling may allow verification of this prediction in the near future . We undertook a careful effort to characterize the global and local spike train statistics through assessing CV , LV , and power spectra . A considerable theoretical literature has been devoted to the significance of neuronal variability and its use to determine the statistical properties of spike trains and their functional relevance [10 , 43 , 44 , 45 , 46] . In particular , the presence of CV values greater than 1 . 0 that is characteristic of random Poisson processes has piqued the interest of theorists , and such values were present in some of our recordings . Previous work has related such high variability to spike initiation non-linearities [46 , 47] or dendritic coincidence detection [46 , 48] , because an integrator over many random inputs would result in a very high degree of regularity in the output . However , our study suggests an alternative mechanism , by which high CV values result from rate covariances in the population of PC inputs to the model neuron . These input rate-covariances lead to constantly changing firing rates in the output as well , which increases the CV . A hallmark of this effect is that local spike train variability of 2 successive ISIs ( LV ) is much less affected and the outcome values of LV are smaller than the CV , unlike in random processes . Therefore , our data and simulations indicate that the assumption of a stationary statistical process underlying neuronal spike trains should be abandoned for the awake condition . Our method of using firing rate templates with specific proportions of co-variance to drive output spiking indeed capture the observed spiking irregularity of data from awake animals well . Parameterizing the degree of these covariances needed to match recorded spike train statistics allowed us to estimate of the required population rate covariance in the behaving animal , and such modeling can therefore shed some light on potential population coding properties in the brain . Another extensive line of theoretical and modeling work has focused on ‘balanced state’ networks , where inhibition and excitation are matched [44 , 49 , 50 , 51] . These recurrent networks of integrate and fire neurons can show Poisson irregularity in firing [50] , and a CV > 1 when co-varying sensory rate fluctuations are transmitted [51] . The required balance between excitation and inhibition in this network state is similar to the balance of excitation and inhibition needed in synaptic input applied to CN neurons with dynamic clamping in order to result in irregular firing patterns with random input spike trains [16 , 22] , a property well replicated in our model [26] . The present study extends this work to the awake state and demonstrates that these concepts fully suffice to explain the spike train statistics recorded in awake alert mice when rate covariance between inputs is added .
Animals . Experiments were performed on male and female adult C57BL/6J ( B6 ) mice ( 18–25 g; The Jackson Laboratory ) . All mice used in this study were raised and all experiments were performed in accordance with procedural guidelines approved by the University of Tennessee Health Science Center Animal Care and Use Committee under protocol # 13–077 . Details about surgical procedures to implant a head post and a recording chamber over the cerebellum were previously published [5 , 52] . During recording mice were head-fixed to a metal holder and the body was loosely covered with a plastic tube to limit body movements . Respiratory behavior was monitored with a thermistor ( Measurement Specialties ) placed in front of one nostril . Breathing cycles were measured as increasing and decreasing temperature changes caused by exhale and inhale movements , respectively . Peaks and troughs in the respiratory signals corresponded to the ends of expiration and inspiration cycles , respectively . Trough times were detected from the analog thermistor output sampled at 1 KHz and used throughout this study as respiratory event markers for respiratory spike train modulation and as alignment for respiratory peri-event histograms . Up to seven recording electrodes ( glass-insulated tungsten/platinum; 80 μm O . D . ; impedance , 3–7 MΩ ) were inserted acutely into the cerebellum during each recording session using a computer-controlled microdrive ( System Eckhorn; Thomas Recording ) . Vermal Purkinje cells were identified by recording depth , a high spontaneous activity rate , and the presence of complex spikes [53] . Mossy fibers were identified using previously described criteria based on granular layer identification and spiking characteristics [54] . Single unit recordings from CN neurons were identified by electrode depth , the electrode passing through an area without spiking activity ( i . e . the white matter embedding the CNs ) before reaching the nucleus , and finally by the presence of sustained spiking ( ~10–70 Hz ) without the occurrence of complex spikes . Recording locations were verified by placing small electrolytic lesions during the last 2 recording days and anatomical reconstruction from 50 μm coronal sections with a cresyl violet staining to align lesion sites with stereotaxic atlas coordinates [55] . Spikes were sorted off-line using Spike2 software ( Cambridge Electronic Design ) and only neurons with a clear refractory period in the ISI histogram and stable spike size over at least 45 s were used for further analysis . This resulted in a data set of 21 PCs , 11 MF , and 16 CN neurons . Spike trains were aligned on respiratory event markers ( end of inspiration ) to create a respiratory PSTH . A confidence interval ( z-score ) to determine significant modulation was constructed by shuffling the respiratory event times 100 times and creating a shuffled PSTH for each instance . Respiratory modulation exceeding the 95% confidence percentile for multiple data points in sequence was deemed significant . The amplitude of modulation was scored by the area under the largest peak or trough of the modulation after baseline subtraction and was scaled to units in spikes , thus yielding a measure of the number of spikes adding or missing in the PSTH peak or trough compared to the shuffle predictor . Using Matlab ( MathWorks , Inc . ) we designed an algorithm to create artificial spike trains ( AST ) that could replicate the observed spike trains statistics and respiratory modulation . We could not directly use experimentally recorded spike trains to drive our CN simulation input because we had at most triple simultaneous PC recordings whereas 50 simultaneous spike trains are needed as input to the model . We used cross correlation and spike covariance analysis [5] to determine the types of cross-correlation and rate covariance present between pairs of simultaneous spike trains and designed an algorithm that could extrapolate these properties to larger spike train populations . Our algorithm uses as core concept the method of rate templates , which are constructed from recorded spike trains by convolving each spike with a Gaussian ( see Supplemental Methods , the full Matlab algorithm is available on ModelDB ) . In the next step of the algorithm we drew gamma distributed spike trains using a mean ISI tracking our rate template and using the shape parameter κ ( kappa ) experimentally determined by the LV from our recordings ( see Results ) . This method allowed us to add respiratory rhythmic modulation in spike trains in a flexible way , by convolving rate templates with a gain-scaled version of the mean peri-stimulus time histogram ( PSTH ) triggered by each cycle of respiration . In order to convolve rate templates and respiratory PSTH functions independent of absolute firing rates , both rate functions were normalized to 1 . 0 before combining them , and the resulting combined rate function was scaled back to the desired mean firing rate of the output AST . Our new method of creating ASTs is quite general and could be used to incorporate any other known rate changes related to behavior . We expect that this method will be of general use in the neural simulation community . In this study we utilized our existing 486 compartment model including the updates to the voltage dependent conductances and synapses described in the Supplemental Information of the original publication and previously shown to replicate CN firing with stochastic synaptic input patterns applied by dynamic clamping [26] . This model has a set of 6 voltage gated and 1 calcium dependent conductance to match the spike shape , spontaneous firing , and responses to depolarization/hyperpolarization of slice CN recordings closely . It also includes 2 inactivating inward conductances that control rebound bursting after strong hyperpolarization [7] . These rebound conductances were present in the model used here , but due to the lack of strong hyperpolarizations with the input patterns constructed to match the waking condition they remained largely inactivated and rebound firing was not observed . In the present study , we included one further model update by incorporating a detailed version of the short term plasticity rules in the PC synapses on CN neurons experimentally determined [20 , 21] . The model depression rule is based upon the rate dependent release probabilities at multiple release sites as estimated by Telgkamp and Raman , 2004 . These STD rules required a re-write of the Genesis 2 . 3 synchan object base code as they could not be achieved with existing synaptic mechanisms in Genesis . The new C base code as well as the updated model definition are available in ModelDB , https://senselab . med . yale . edu/modeldb/ShowModel . cshtml ? model=229279 . Synaptic inputs were modeled as a dual exponential alpha function with rise and decay time constants matching voltage clamp recordings in slices ( see detailed explanation in supplemental materials , [7] ) . Each spike from our PC ASTs triggered a unitary IPSC with a peak amplitude controlled by the Gin parameter . The range of Gin used was between 2 and 20 nS for parameter scans , and values of 4 or 16 nS were used in most simulation runs exploring respiratory rate modulations . This compares to an average IPSC size of 9 . 4 nS with minimal ( single axon ) stimulation in slices and an observed experimental range of 1–25 nS . Our excitatory MF inputs triggered both an NMDA and an AMPA EPSC , as mixed currents have been found in experiments [40 , 41 , 56] . 48 inhibitory and 48 excitatory synapses were distributed randomly across the 485 dendritic compartments , and 2 inhibitory synapses were placed on the soma . There was no somatic excitation as excitatory synapses are not observed on the soma of CN neurons [57 , 58] . Each synapse was connected to one AST input spike train . Simulations were run in batches on a Linux cluster , where each batch completed a matrix of parameter settings . All simulations were run to produce 115s of output data . Binary and spike event output files from simulation batches were put into a Pandora database format and analyzed with custom made Matlab scripts . Power spectra were determined using functions from the Chronux Matlab toolbox ( http://chronux . org/ ) .
|
Detailed computer simulations of biological neurons can make an important contribution to our understanding of how the brain works . In this paper we use such a model of a neuron that represents the output from the cerebellum . We can show that the inhibition this neuron type receives from Purkinje cells in the cerebellar cortex is well suited to pass a detailed time course of movement control to the output of the cerebellum . Importantly we find that this type of coding requires a population of Purkinje cells that pass the same temporal coding of spike rate to the output neurons in the cerebellar nuclei .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"nervous",
"system",
"membrane",
"potential",
"brain",
"electrophysiology",
"random",
"variables",
"neuroscience",
"covariance",
"mathematics",
"computational",
"neuroscience",
"cerebellum",
"coding",
"mechanisms",
"cerebral",
"cortex",
"animal",
"cells",
"probability",
"theory",
"cellular",
"neuroscience",
"cerebellar",
"nuclei",
"cell",
"biology",
"anatomy",
"synapses",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"computational",
"biology",
"neurophysiology",
"purkinje",
"cells"
] |
2017
|
Robust transmission of rate coding in the inhibitory Purkinje cell to cerebellar nuclei pathway in awake mice
|
Low vitamin D levels in human immunodeficiency virus type-1 ( HIV ) infected persons are associated with more rapid disease progression and increased risk for Mycobacterium tuberculosis infection . We have previously shown that 1α , 25-dihydroxycholecalciferol ( 1 , 25D3 ) , the active form of vitamin D , inhibits HIV replication in human macrophages through the induction of autophagy . In this study , we report that physiological concentrations of 1 , 25D3 induce the production of the human cathelicidin microbial peptide ( CAMP ) and autophagic flux in HIV and M . tuberculosis co-infected human macrophages which inhibits mycobacterial growth and the replication of HIV . Using RNA interference for Beclin-1 and the autophagy-related 5 homologue , combined with the chemical inhibitors of autophagic flux , bafilomycin A1 , an inhibitor of autophagosome-lysosome fusion and subsequent acidification , and SID 26681509 an inhibitor of the lysosome hydrolase cathepsin L , we show that the 1 , 25D3-mediated inhibition of HIV replication and mycobacterial growth during single infection or dual infection is dependent not only upon the induction of autophagy , but also through phagosomal maturation . Moreover , through the use of RNA interference for CAMP , we demonstrate that cathelicidin is essential for the 1 , 25D3 induced autophagic flux and inhibition of HIV replication and mycobacterial growth . The present findings provide a biological explanation for the benefits and importance of vitamin D sufficiency in HIV and M . tuberculosis-infected persons , and provide new insights into novel approaches to prevent and treat HIV infection and related opportunistic infections .
Human immunodeficiency virus type-1 ( HIV ) is a global health problem that has infected 60 million people and caused 25 million deaths worldwide . Currently , there are an estimated 33 million people living with HIV including 2 million children . Despite the immune defense mechanisms that the host deploys against HIV and improved antiretroviral therapies , the virus persists in long lived cells including resting T cells , macrophages and dendritic cells . One-third of HIV-infected individuals are co-infected with Mycobacterium tuberculosis , a leading cause of death among people living with HIV . It has been proposed that the increase in M . tuberculosis pathology associated with HIV infection is caused by the disruption of the local immune response within the tuberculosis granulomas , decreasing their ability to contain M . tuberculosis leading to increased mycobacterial replication , dissemination and clinical disease [1]–[5] . Several studies have linked vitamin D deficiency ( 25-hydroxycholecalciferol ( 25D3 ) deficiency ) with an increased risk for susceptibility to tuberculosis and active disease both in the presence [6] and absence of HIV infection [7]–[14] . Few studies have examined the association between vitamin D status and HIV disease progression and survival . However , the data available suggest that HIV-infected individuals have lower levels of 25D3 and/or the vitamin D3 active metabolite , 1α , 25-dihydroxycholecalciferol ( 1 , 25D3 ) than uninfected individuals [15]–[21] with the lowest concentrations found in persons with AIDS [19] , [20] . Additionally , women with low levels of 25D3 have an increased risk of HIV disease progression [22] and infants born to HIV-infected mothers with low 25D3 levels have an increased risk of HIV infection and increased mortality [23] . Although 25D3 has no direct anti-mycobacterial or antiretroviral effect , its hormonally active form , 1 , 25D3 , modulates the immune response and has been shown to exert both anti-mycobacterial [24]–[27] and anti-HIV effects [27] , [28] in vitro . Macroautophagy ( herein referred to as autophagy ) is a trafficking pathway whereby cytoplasmic constituents such as sub-cellular organelles and microbial pathogens are engulfed by autophagosomes which fuse with lysosomes , forming autolysosomes , degrading the engulfed components . As an obligatory intracellular parasite , HIV survival is dependent upon its ability to exploit host cell machinery for replication and dissemination and to circumvent cellular processes that prevent its growth . During infection , HIV down regulates Beclin-1 and microtubule-associated protein 1 light chain 3B ( LC3B ) -II , reducing both basal autophagy and the numbers of autophagosomes per cell [29] , [30] . However , silencing of autophagy proteins inhibits HIV infection of HeLa cells [31] and macrophages [28] . M . tuberculosis interferes with the biogenesis of phagolysosomes , and persists and replicates in macrophages within special immature phagosomes characterized by the exclusion of the vacuolar H+ ATPase and the absence of lysosomal hydrolases . In this study , we investigated the effect of 1 , 25D3 on productive HIV and M . tuberculosis infection of macrophages . We demonstrate that 1 , 25D3 inhibits HIV replication and mycobacterial growth using autophagic machinery .
Previous studies have demonstrated that physiological concentrations of 1 , 25D3 have indirect antimicrobial activity against M . tuberculosis and HIV . However , to date , no study has assessed the ability of physiological levels of 1 , 25D3 to inhibit HIV during M . tuberculosis co-infection . Therefore , we initially assessed whether 1 , 25D3 inhibits HIV replication in macrophages by comparing the extent to which 1 , 25D3 pre-treatment affects HIV p24 antigen accumulation in the supernatants of macrophages that were subsequently infected with HIV and/or M . tuberculosis . 1 , 25D3 induced a dose-dependent inhibition of HIV replication with 50 pmol/L being the minimum concentration required to significantly inhibit HIV by day 7 ( 68% reduction; P = 0 . 003; Figure 1A ) , while 100 pmol/L , the concentration found in healthy 25D3-sufficient HIV-uninfected plasma , inhibited HIV by 82% ( P<0 . 001; Figure 1A ) . Moreover , 1 , 25D3 induced a dose-dependent reduction in cells expressing HIV p17 ( Figure 1A ) . In the presence of M . tuberculosis co-infection , 1 , 25D3 also induced a dose-dependent inhibition of HIV replication . The minimum dose of 1 , 25D3 required to significantly inhibit HIV replication did not change ( 54% reduction; P = 0 . 005; Figure 1A ) and the profile of intracellular HIV p17 expression post-1 , 25D3 treatment was similar to M . tuberculosis-unexposed cells ( Figure 1A ) . To evaluate the capacity of 1 , 25D3 to suppress the mycobacterial growth in HIV-infected macrophages , MDM were infected with M . tuberculosis H37Rv at infection ratios ( M . tuberculosis∶target cell ) of 8∶1 . Cells were then cultured for 7 d in the presence of varying concentrations of 1 , 25D3 after which they were lysed and the number of colony forming units ( cfu ) of M . tuberculosis in macrophages enumerated . There were no statistically differences observed in cfu at day 0 . However , at day 7 , 1 , 25D3 induced a dose-dependent reduction in cfu counts that became significant at 50 pmol/L ( P<0 . 001; Figure 1B ) . This observed effect was dependent upon macrophage infection by M . tuberculosis as 1 , 25D3 had no effect on mycobacterial viability when grown in cell culture media or Middlebrook 7H9 broth alone ( data not shown ) . Moreover , 1 , 25D3 also induced a dose-dependent reduction in M . tuberculosis-positive cells as assessed by flow cytometry ( Figure 1B ) . The effect of HIV infection on the capacity of macrophages to limit mycobacterial growth was next assessed . Although the presence of HIV resulted in a slightly increased cfu count at day 7 , there were no significant differences both in cfu count or in growth index ( cfu day 7∶cfu day 0 ) compared with M . tuberculosis only treated cells . In the presence of HIV infection , 1 , 25D3 treatment was associated with a dose-dependent reduction in mycobacterial viability that became statistically significant at 50 pmol/L , at which point the cfu count was less than at day 0 ( 8 . 9×104 versus 7 . 5×104 cfu/well; P = 0 . 001; Figure 1B ) . Flow cytometry also revealed that 100 pmol/L 1 , 25D3 significantly inhibited mycobacterial growth ( Figure 1B ) . Dilutions of the lysates were also run in a Mycobacteria Growth Indicator Tube 960 apparatus ( MGIT 960 ) . 1 , 25D3 induced a dose-dependent increase in time to positive ( 75 growth units ) both in the presence and in the absence of HIV infection indicating a smaller initial mycobacterial count post 1 , 25D3 treatments ( Figure 1C ) . It has previously been demonstrated that 1 , 25D3 induces autophagy in human macrophages during infection with either HIV or M . tuberculosis . However , no study has investigated the effect of 1 , 25D3 on autophagy during co-infection . Therefore , the ability of 1 , 25D3 to induce autophagy in HIV and M . tuberculosis co-infected human macrophages was assessed . An established molecular marker for the induction of autophagy is the degree of LC3B lipidation [32] . During autophagy , cytosolic LC3B-I is converted to LC3B-II by a ubiquitin-like system that involves autophagy related protein-7 ( ATG7 ) , ATG3 and the ATG5-ATG12 complex . The ATG5-ATG12 complex ligates LC3B-II to the nascent autophagosome membrane through phosphatidylethanolamine with the LC3B-II associated with the inner membrane degraded after fusion of the autophagosome with lysosomes . Therefore , the conversion of LC3B-I to LC3B-II and its turnover is an indicator of autophagy induction and flux [32] . 1 , 25D3 treatment of MDM induced an increase in LC3B-II in cells that were incubated with HIV and/or M . tuberculosis ( Figure 2A ) . The accumulation of LC3B-II was increased in the presence of the lysosomal protease inhibitor pepstatin A regardless of infection status ( Figure 2B ) , indicative of autophagic flux [33] . When autophagosomes are formed , LC3B redistributes from a soluble diffuse cytosolic pattern to an insoluble autophagosome-associated vacuolar pattern [34] , [35] allowing the quantification of autophagosome-associated LC3B-II in human macrophages using saponin resistance and flow cytometry [35] . Staining for endogenous LC3B in saponin washed macrophages revealed that the percentage of cells containing a saponin resistant fraction significantly increased with dose of 1 , 25D3 ( P<0 . 05; Figure 2C ) . In uninfected cells and in HIV and/or M . tuberculosis infected cells , co-treatment with pepstatin A significantly increased the percentage and fluorescent signal of cells containing a saponin resistant fraction upon treatment with 100 pmol/L 1 , 25D3 indicative of autophagic flux ( P<0 . 05; Figure 2D ) . Another control to confirm that the increase in LC3B-II observed in 1 , 25D3 treated cells during HIV and/or M . tuberculosis infection represents increased autophagic flux , rather than an accumulation of LC3-positive autophagosomes is the measurement of polyubiquitin-binding protein p62 ( sequestosome 1 ) degradation through immunoblotting as p62 binds LC3B [33] . Inhibition of autophagy leads to an increase in p62 protein levels while p62- and LC3-positive bodies are degraded in autolysosomes during autophagic flux [36] . HIV and/or M . tuberculosis infected MDM were subjected to 1 , 25D3 stimulus for 7 d after which p62 protein levels were measured by western blot analysis . 1 , 25D3 treatment induced a decrease in p62 protein levels in HIV and/or M . tuberculosis infected MDM compared to vehicle-treated cells , corresponding to a stimulation of autophagic flux ( Figure 2E ) . Although there was an increase in autophagic markers in the absence of visible pyknosis , karyorrhexis , or plasma membrane blebbing , it was important to confirm that the cells were not undergoing cell death at the physiological concentrations being used , as the induction of excessive autophagy can cause cell death in mammalian cells in experimental systems in vitro [37] . Therefore , plasma membrane breakdown ( as a sign of cytotoxicity ) using the lactate dehydrogenase assay was measured in combination with the WST-1 assay that measures the activity of the mitochondrial respiratory chain ( as an indicator of viable cells ) . After 7 d at physiological concentrations , 1 , 25D3 exhibited no cytotoxic effects in the presence or absence of HIV and/or M . tuberculosis infection ( P>0 . 05; Figure 2F ) . Recent studies have demonstrated that physiological levels of 1 , 25D3 inhibit M . tuberculosis [38] and HIV [28] through autophagy dependent mechanisms . Therefore , it was important to investigate the relationship between infection status and the 1 , 25D3-mediated induction of autophagy . At 7 d post- HIV infection , the majority of MDM were positive for HIV p17 ( Figure 3A ) . Treatment with 100 pmol/L 1 , 25D3 , reduced the number of cells with detectable HIV p17 and was accompanied by an overall reduction in mean HIV p17 antibody fluorescence that was greatest in cells that also exhibited saponin resistant LC3B-II ( Figure 3A ) . In the absence of 1 , 25D3 and HIV infection , incubation of MDM with M . tuberculosis for 7 days resulted in 75% of MDM becoming positive for M . tuberculosis , one third of which were also positive for saponin resistant LC3B-II ( Figure 3B ) . All M . tuberculosis negative cells were saponin resistant LC3B-II negative . 100 pmol/L 1 , 25D3 significantly decreased the number of M . tuberculosis positive cells ( P = 0 . 002; Figure 3B ) while concomitantly increasing the number of saponin resistant LC3B-II positive cells . Co-culture of MDM with both HIV and M . tuberculosis for 7 days resulted in the inhibition of saponin resistant LC3B-II that appeared in M . tuberculosis-only infected MDM . Similar to HIV and M . tuberculosis-only infections , incubation with 100 pmol/L 1 , 25D3 increased the numbers of saponin resistant LC3B-II positive cells while reducing the number of M . tuberculosis positive but not HIV p17 positive cells ( Figure 3C and 3D ) . Cells co-infected with HIV/M . tuberculosis demonstrated saponin resistant LC3B-II when treated with 50 pmol/L 1 , 25 D3 , while cells infected with HIV did not show saponin resistant LC3B-II when treated with 50 pmol/L 1 , 25 D3 . ( Figure 3D ) . The contribution of 1 , 25D3-induced autophagy in 1 , 25D3-mediated inhibition of HIV and mycobacterial growth was investigated by inhibiting sequential steps of the autophagy pathway . As the physical interaction of class III phosphatidylinositol 3-kinase with Beclin-1 forms the phosphatidylinositol 3-kinase class III kinase complex and this complex is essential for the induction of autophagosome formation at the vesicle elongation step , RNAi for Beclin-1 was initially employed . Concomitant with the findings that HIV utilizes autophagic machinery for replication [28] , [31] , [39] , the supernatant p24 antigen concentration in Beclin-1 silenced cells was decreased in the absence of 1 , 25D3 regardless of M . tuberculosis infection status . In the absence of M . tuberculosis infection , Beclin-1 silencing ( Figure 4A ) significantly reversed the 1 , 25D3 mediated inhibition of HIV at day 7 from 91% to 44% ( P<0 . 001; Figure 4B ) . Conversely , in the presence of M . tuberculosis infection , Beclin-1 silencing reduced 1 , 25D3 inhibition of HIV infection from 88% to 15% ( P<0 . 001; Figure 4B ) . In the absence of HIV , Beclin-1 silencing increased mycobacterial growth 51% at day 7 increase compared with the scrambled non-target control RNAi treated cells ( P = 0 . 0006 ) . HIV co-infection increased this difference to 55% ( P<0 . 001 ) although the difference between the cfu counts between the M . tuberculosis and HIV/M . tuberculosis treated Beclin-1 RNAi transduced cells was not significant ( P = 0 . 36 ) . The effect of 1 , 25D3 on the capacity of macrophages to limit mycobacterial growth was next assessed . Beclin-1 silencing protected against the anti-mycobacterial effects of 1 , 25D3 both in the presence and in the absence of HIV infection ( P>0 . 05; Figure 4C ) . A similar profile was observed using the MGIT 960 ( Figure 4D ) . During autophagy , cytosolic LC3B-I is converted to LC3B-II by an ubiquitin-like system that involves the autophagy-related 7 ( ATG7 ) homologue , ATG3 and the ATG5-ATG12 complex . The ATG5-ATG12 complex ligates LC3B-II to the nascent autophagosome membrane through phosphatidylethanolamine . Therefore , RNAi of ATG5 inhibits autophagosome formation . In agreement with the above findings that HIV was inhibited through Beclin-1 silencing , ATG5 silencing ( Figure 5A ) also had an inhibitory effect on HIV replication alone and in the presence of M . tuberculosis co-infection ( Figure 5B ) . In the absence of M . tuberculosis co-infection , ATG5 RNAi completely abrogated the 1 , 25D3 mediated inhibition of HIV by day 7 ( P<0 . 001; Figure 5B ) . In the presence of M . tuberculosis co-infection , ATG5 silencing reduced the 1 , 25D3 mediated inhibition of HIV at day 7 from 91% to 4% ( P<0 . 001; Figure 5B ) . As for Beclin-1 RNAi , M . tuberculosis growth was significantly increased in HIV infected and uninfected ATG5 silenced cells in the absence of 1 , 25D3 treatment ( P<0 . 003; Figure 5C ) . Moreover , the effect of 1 , 25D3 on the capacity of macrophages to limit mycobacterial growth was severely diminished both in the presence and absence of HIV infection ( P<0 . 001; Figure 5C ) . A similar profile was observed using the MGIT 960 . ATG5 RNAi abrogated the 1 , 25D3 inhibition of mycobacterial growth both in the absence and presence of HIV infection ( Figure 5D ) . We next investigated whether autophagosome acidification , a late stage event during autophagy , is required for the 1 , 25D3-mediated autophagic inhibition of HIV and M . tuberculosis . During autophagy , lysosomes fuse with autophagosomes to form autophagolysosomes . Macrophages were treated with bafilomycin A1 , an inhibitor of the vacuolar H+ ATPase and thus autophagosome-lysosome fusion , and subsequently infected with HIV and/or M . tuberculosis . Bafilomycin A1 modestly increased HIV production over 7 d in the absence of 1 , 25D3 and M . tuberculosis infection although this was not significant . It had no effect on HIV replication in the presence of M . tuberculosis infection . In the presence of 1 , 25D3 , bafilomycin A1 abrogated the 1 , 25D3-mediated inhibition of HIV both alone and in the presence of M . tuberculosis ( Figure 6A ) . Mycobacterial growth was similarly increased after bafilomycin A1 treatment in the absence of 1 , 25D3 in both infection models although this was not significant . Bafilomycin A1 significantly reduced the inhibitory effect of 100 pmol/L 1 , 25D3 both in the presence ( P = 0 . 05; Figure 6B ) and absence of HIV co-infection ( P = 0 . 06; Figure 6B ) . We obtained comparable results using the MGIT 960 with 100 pmol/L 1 , 25D3 significantly inhibiting mycobacterial growth both in the presence and absence of HIV infection ( P<0 . 0001; Figure 6C ) . These results suggest that the acidic pH of autophagolysosomes is required for the autophagy-mediated control of both HIV replication and M . tuberculosis growth by 1 , 25D3 . After lysosomes fuse with autophagosomes to form autophagolysosomes , the sequestered components are then degraded by lysosomal hydrolases and presumably released into the cytosol by lysosomal efflux permeases . Therefore the effect of lysosomal hydrolases in 1 , 25D3-mediated inhibition of HIV and M . tuberculosis through autophagy was examined using SID 26681509 , a novel thiocarbazate specific inhibitor of the lysosome hydrolase cathepsin L . In the absence of 1 , 25D3 there was no net inhibition of either HIV or M . tuberculosis growth ( Figure 7A ) . Moreover , in the presence of 1 , 25D3 , SID 26681509 abrogated the inhibition of both HIV and M . tuberculosis ( Figure 7B and 7C ) . Together , these data indicate that the 1 , 25D3-mediated induction of autophagy in macrophages inhibits HIV and M . tuberculosis replication regardless of co-infection status . Previous studies have demonstrated that the human cathelicidin microbial peptide ( CAMP ) is required for the 1 , 25D3 mediated antimycobacterial activity against M . tuberculosis [38] , [40] and the 1 , 25D3-mediated autophagy in human macrophages [38] . To address the role of CAMP in 1 , 25D3-induced antimicrobial activity , RNAi for CAMP was employed . Transduction of shCAMP into MDM completely silenced the 1 , 25D3-induced expression of CAMP during HIV and M . tuberculosis co-infection ( Figure 8A ) . Moreover , CAMP silencing markedly inhibited 1 , 25D3-induced autophagy , whereas MDM transduced with a scrambled ( shNS ) showed increased saponin resistant LC3 , consistent with autophagosome formation . Concomitant with our findings that autophagy is required for the restriction of HIV replication , CAMP silencing reduced 1 , 25D3 inhibition of HIV both alone ( 90% shNS versus 10% shCAMP; P<0 . 001; Figure 8C ) and in the presence of M . tuberculosis infection ( 91% shNS versus 10% shCAMP; P<0 . 001; Figure 8C ) . Moreover , CAMP silencing completely blocked the 1 , 25D3-mediated antimycobacterial activity ( P<0 . 001; Figure 8D ) . These data suggest that CAMP is required for 1 , 25D3-mediated antimicrobial activity against both intracellular M . tuberculosis and HIV in human MDM .
Despite much progress in the care and treatment of persons infected with HIV and those suffering from tuberculosis , the two pathogens are inextricably linked as important causes of morbidity and mortality worldwide . Thus , the effective treatment of persons co-infected with HIV and M . tuberculosis remains a major unmet challenge . In the case of tuberculosis , administration of vitamin D and/or sunlight exposure has a long and colorful history [11] . In fact , the beneficial effects of cod liver oil in the treatment of tuberculosis were first recognized in 1849 and Niels Finsen received the Nobel Prize in 1903 for his discovery of ultra-violet light as an effective therapy for the cutaneous form of tuberculosis , lupus vulgaris . However , more than 100 years later , the therapeutic benefits of vitamin D and the plasma levels of 25D3 required to provide clinical benefit remain controversial [41] , [42] . Our results demonstrate that physiological concentrations of 1 , 25D3 inhibit both HIV and M . tuberculosis replication in human macrophages through an autophagy and CAMP dependent mechanism , and that 1 , 25D3 can act as a potent stimulator of innate antimicrobial responses . The ability of macrophages to kill intracellular pathogens is pivotal to the outcome of microbial infections . Within macrophages , M . tuberculosis and HIV reside in phagosomes and evade host microbial mechanisms by blocking phagosome maturation and fusion with lysosomes [39] , [43] . However , the host can overcome this block through the induction of autophagy [28] , [44] , [45] . Previous studies have shown that physiological concentrations of 1 , 25D3 mediate the induction of autophagy and the individual killing of M . tuberculosis [38] and HIV [28] . To our knowledge the present study is the first to demonstrate that 1 , 25D3 inhibits both M . tuberculosis and HIV co-infection of macrophages through the induction of autophagy mediated by CAMP . The in vitro findings reported here are supported by association studies that have linked low levels of 25D3 and/or 1 , 25D3 with increased risk of , or severity of infection with HIV [15]–[23] and M . tuberculosis [7]–[10] , [12]–[14] , [21] . Why HIV-infected individuals tend to have lower levels of 1 , 25D3 and/or 25D3 is largely unknown but is thought to be related to inadequate renal 1α-hydroxylation mediated by pro-inflammatory cytokines and/or a direct effect of antiretroviral drugs [16] , [19] , [46] . The effects of HIV viral products on 1 , 25D3 and/or 25D3 syntheses have not been evaluated . Vitamin D deficiency is conservatively defined by most experts as <50 nmol/L 25D3 [47]; 52–72 nmol/L 25D3 is considered to indicate insufficiency and >73 nmol/L considered sufficient [47] . In contrast to this , the estimated mean concentration of 25D3 present in people worldwide is just 54 nmol/L [48] . Four genes are known to contribute to the variability of serum 25D3 concentrations: 7-dehydrocholesterol reductase ( involved in cholesterol synthesis and the availability of 7-dehydrocholesterol in the skin ) , 25-hydroxylase CYP2R1 ( cytochrome P450 , family 2 , subfamily R , polypeptide 1 ) and CYP24A1 ( cytochrome P450 , family 24 , subfamily A , polypeptide 1 ) ( degrades and recycles 1 , 25D3 ) and GC ( group-specific component [vitamin D binding protein] ) which encodes for the vitamin D binding protein . Genetic variations at these loci were recently identified to be significantly associated with an increased risk of 25D3 insufficiency [49] . A better understanding of the pathogenesis of tuberculosis will be necessary if novel interventions are to be developed to effectively treat persons infected with M . tuberculosis and HIV . The induction of autophagy to enhance treatment is attractive for a number reasons including: 1 ) autophagy would work at the cellular level to improve intracellular killing of both pathogens; 2 ) enhanced autophagy is likely to be equally active against multi-drug resistant ( MDR ) and extensively drug resistant ( XDR ) M . tuberculosis as it is against drug sensitive M . tuberculosis; 3 ) autophagy has the potential to inhibit non-replicating HIV and M . tuberculosis within endosomes; 4 ) autophagy promotes innate and adaptive immunity; and 5 ) HIV or M . tuberculosis resistance is unlikely to develop . The characterization of the 1 , 25D3 mediated antimicrobial mechanism in macrophages provides further evidence of the link between vitamin D and the immune system . The intracrine nature of this mechanism suggests that the ability of 25D3 to promote M . tuberculosis and HIV killing could be affected by the efficiency of the synthesis of 1 , 25D3 by macrophages . Unlike the parathyroid-hormone responsiveness of renal cytochrome P450 , family 27 , subfamily B , polypeptide 1 ( CYP27B1 ) , extra-renal CYP27B1 is not subject to the same feedback control so that the local synthesis of 1 , 25D3 in macrophages probably reflects the availability of 25D3 . Toll-like receptor 2/1 agonists and interferon gamma upregulate the expression of the CYP27B1 which 1α-hydroxylates 25D3 into 1 , 25D3 , activating and upregulating the expression of the vitamin D ( 1 , 25D3 ) receptor ( VDR ) leading to the induction of CAMP and autophagic flux [50] , [51] . When serum is 25D3 deficient , TLR2/1 agonists ( <25 nmol/L ) [50] and interferon gamma ( <45 nmol/L ) [51] are unable to induce the expression of CAMP from monocytes or macrophage . In the present study we demonstrate that the 1 , 25D3-induced autophagy and anti-mycobacterial activity was dependent on the expression of endogenous CAMP in agreement with previous findings [38] . Moreover , by using RNAi for CAMP we demonstrate , for the first time , that the 1 , 25D3 induced autophagy and subsequent autophagy-induced antiretroviral activity is dependent upon endogenous CAMP expression . This finding highlights the importance of endogenous CAMP during autophagy induction and antiretroviral activity . High concentrations of exogenous synthetic cleaved form of CAMP ( LL-37 ) have been shown to reduce mycobacterial growth [25] , [38] , [40] and HIV replication [28] . However , at the much lower concentrations secreted in vitro by 1 , 25D3 treated macrophages , exogenously added LL-37 has not been shown to have an inhibitory effect on either organism [25] , [28] or induce the formation of autophagosomes [28] . However , Yuk et al . [38] clearly showed that CAMP is upstream of Beclin-1 and ATG5 induction post-1 , 25D3 treatment and we show that in CAMP silenced cells 1 , 25D3 fails to induce LC3B lipidation . Further work is necessary to determine the precise role CAMP has in 1 , 25D3 induced autophagy and antimicrobial activity . When modeling the replication of mycobacteria in human cells , we used the standard protocol of incubating macrophages with M . tuberculosis followed by washing the cells of non-phagocytosed bacilli . In this way , the intracellular replication of mycobacteria in macrophages can be assessed . However , M . tuberculosis is able to replicate extracellularly in vivo and as would be expected , we observed that 1 , 25D3 had no effect on M . tuberculosis survival in the extracellular compartment . In addition , as the cfu assay is cumbersome for high-throughput analyses , we adopted and ran samples in the MGIT 960 using a standard curve derived from the untreated macrophage cell lysates and found that the results obtained with this method were closely aligned with those obtained using the more subjective cfu method . This procedure can be used to more rapidly and efficiently screen compounds for the ability to inhibit M . tuberculosis replication compared to the standard assay . In summary , this study establishes a role for autophagy during the early phases of HIV infection and demonstrates that the induction of autophagy by 1 , 25D3 can inhibit HIV and M . tuberculosis co-infection in macrophages . Well-controlled clinical trials are needed to determine if vitamin D supplementation is of value for prevention or as adjunctive treatment in HIV-infected persons against active tuberculosis . Dissecting the molecular mechanisms by which HIV and M . tuberculosis utilize autophagy has the potential to lead to the identification of novel drug candidates that can be used to prevent and treat HIV infection and related opportunistic infections including tuberculosis .
Venous blood was drawn from HIV seronegative subjects using a protocol that was reviewed and approved by the Human Research Protections Program of the University of California , San Diego ( Project 08-1613 ) in accordance with the requirements of the Code of Federal Regulations on the Protection of Human Subjects ( 45 CFR 46 and 21 CFR 50 and 56 ) . All human studies were conducted according to the principles expressed in the Declaration of Helsinki . Written informed consent was obtained from all study participants prior to their participation . Peripheral blood mononuclear cells were isolated from whole blood by density gradient centrifugation over Ficoll-Paque Plus ( GE Healthcare ) . Cells were then incubated overnight at 37°C , 5% CO2 in RPMI 1640 ( Gibco ) supplemented with 10% ( v/v ) charcoal/dextran treated , heat-inactivated fetal bovine serum ( FBS; Gemini Bio-Products ) and 10 ng/mL macrophage colony stimulating factor ( R&D Systems ) , after which non-adherent cells were removed by aspiration . The same lot of FBS was used throughout all experiments . Monocyte derived macrophages were obtained by incubating the adherent population in the culture medium at 37°C , 5% CO2 for a further 10 days . The recovered cells were >95% CD163+ , as determined by flow cytometry . 1 , 25D3 , pepstatin A , bafilomycin A1 and SID 26681509 were purchased from Sigma . Bafilomycin A1 and SID 26681509 were used at 100 and 50 nmol/L , respectively with pretreatment for 1 h before addition of 1 , 25D3 or vehicle control . Cytotoxicity of the different chemicals at the concentrations used was tested by the trypan blue dye exclusion assay , and none was found to be cytotoxic ( viability was >99% ) . MISSION small hairpin ( shRNA ) lentiviral particles were obtained from Sigma . Lentiviral transduction of MDMs with particles for shRNAs targeting ATG5 ( SHCLNV-NM_004849/TRCN0000150940 ) , Beclin-1 ( SHCLNV-NM_003766/TRCN0000033551 ) , CAMP ( SHCLNV-NM_004345/TRCN0000118645 ) , or scrambled non-target negative control ( Scr , SHC002V ) was performed according to the manufacturer's protocol . Electrophoresis and immunoblotting was as previously described [28] . HIVBa-L was obtained through the AIDS Research and Reference Reagent Program , from Suzanne Gartner and Robert Gallo [52] , [53] . Virus stocks were prepared and titered as previously described using the Alliance HIV p24 antigen enzyme-linked immunosorbent assay kit ( ELISA; Perkin Elmer ) [54] . M . tuberculosis H37Rv was kindly provided by Janice Kaping , University of California San Diego . Stock strains were grown for 7–10 days to reach mid-exponential growth phase in Middlebrook 7H9 broth supplemented with 1% glycerol , 0 . 05% polybrene 80 and 10% OADC ( oleic acid , albumin , dextrose , catalase ) enrichment ( all Sigma ) at 37°C . Mycobacterial cultures were pelleted at 3000× g for 10 min and resuspended in Middlebrook 7H9 broth . Clumped mycobacteria were dispersed using ultrasound waves ( 3 to 5 min , 40 kHz; NeyTech ) . The sample was centrifuged at 200× g for 10 min to pellet the clumped bacilli , and the upper mycobacterial suspension was used in all experiments . Cultures were plated for viable colony forming unit ( cfu ) counts on Middlebrook 7H10 agar with OADC enrichment ( BD Diagnostic Systems ) . To rule out the influence of lipopolysaccharide ( LPS ) in the assays , the mycobacterial suspensions were tested by the Limulus amebocyte lysate assay ( Lonza ) . The effective LPS concentration was <2 pg/mL in experiments with mycobacteria to cell ratios of 8∶1 . 5×105 MDM were treated for 4 h with 1 , 25D3 then infected with 105 TCID50 HIVBa-L and/or 4×106 M . tuberculosis for 4 h . After an incubation period of 4 h , noningested M . tuberculosis were removed by washing four times with DPBS . The majority of extracellular mycobacteria ( >99% ) were removed with this process as determined by auramine-rhodamine staining . Each well was repleted with 1 mL RPMI 1640 supplemented with 10% ( v/v ) heat-inactivated FBS with or without 1 , 25D3 , and incubated for 30 min ( time 0 ) and 7 d . At day 4 , 500 µL cell supernatant was sampled a replaced with 500 µL fresh media ±1 , 25D3 added at this point . For RNAi experiments , 1×105 MDM were used and volumes and infectious particles adjusted accordingly . At the end of the incubation periods cells and supernatants were harvested and stored frozen at −70°C until further assessment in the p24 ELISA or the cfu assay . To assess the number of intracellular mycobacteria , cells were thawed and lysed with 0 . 5% ( w/v ) sodium dodecyl sulfate ( Sigma ) . Lysates of infected cells were resuspended vigorously , transferred into screw caps and sonicated in a preheated ( 37°C ) water bath sonicator for 5 min ( 40 kHz; NeyTech ) . Aliquots of the lysates were diluted in Middlebrook 7H9 broth and serial dilutions of each sample were plated in quadruplicate on Middlebrook 7H10 agar with OADC enrichment ( BD Diagnostics ) and incubated at 37°C and 5% CO2 for 30 days . CFU were enumerated using a stereomicroscope . The results are expressed as mean ± SE of CFU/well . Alternatively , a dilution of each sample was placed into a Bactec MGIT 960 ( BD Diagnostics ) . Before inoculation , Bactec MGIT 960 tubes were supplemented as described by the manufacturer ( MGIT [7-ml] package insert; Becton Dickinson ) . The lysates were diluted 1∶40 in Middlebrook 7H9 and a 50 µL aliquot of this was further diluted 1∶10 in Middlebrook 7H9 and inoculated into the MGIT tube ( 500 µL added ) . The tubes were introduced into the Bactec MGIT 960 instrument , as recommended by the manufacturer and incubated either until they were found to be positive by the instrument ( 75 Growth Units ) or for 6 weeks . Results are expressed as mean time to positivity . Aliquots of the cell supernatants were taken for HIV p24 analysis by ELISA at days 0 , 4 and 7 . At all time points , an aliquot of unlysed , infected cells was harvested , enumerated and viability assessed using the trypan blue exclusion assay . Recovery of the cells was >90% in all experiments , with cell viability exceeding 95% at all time points and treatments . Two-tailed , Student's t tests , α = 0 . 05 , were used to assess whether the means of two normally distributed groups differed significantly .
|
Macroautophagy ( autophagy - ‘self-eating’ , lysosome-dependent degradation and recycling of the intracellular components in response to stress ) is an important host defense mechanism against viral and mycobacterial infections . Recent studies have described that activation of autophagy in macrophages reduces the viability of Mycobacterium tuberculosis and HIV due to an intimate autophagy-phagocytosis interaction . Low serum levels of the 25-hydroxycholecalciferol form of vitamin D have been associated with an increased risk for active tuberculosis and HIV disease progression as well as M . tuberculosis susceptibility . In this study , we report that the active form of vitamin D , 1α , 25-dihydroxycholecalciferol inhibits the replication of HIV and M . tuberculosis in a concentration dependent manner . Moreover , by inhibiting key stages in the autophagy pathway , we demonstrate that the inhibition of HIV and mycobacterial growth during single infection or dual infection is dependent not only upon the induction of autophagy , but also through phagosomal maturation . Furthermore , through the use of RNA interference for the human cathelicidin microbial peptide we demonstrate that cathelicidin is essential for the 1α , 25-dihydroxycholecalciferol induced autophagic flux and inhibition of HIV replication and mycobacterial growth . These findings suggest that the induction of autophagy has the potential to be useful in the treatment of persons co-infected with HIV and M . tuberculosis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"immunity",
"tuberculosis",
"innate",
"immunity",
"hiv",
"retrovirology",
"and",
"hiv",
"immunopathogenesis",
"immunology",
"biology",
"viral",
"diseases"
] |
2012
|
Vitamin D Inhibits Human Immunodeficiency Virus Type 1 and Mycobacterium tuberculosis Infection in Macrophages through the Induction of Autophagy
|
Schistosomiasis continues to be an important cause of parasitic morbidity and mortality world-wide . Determining the molecular mechanisms regulating the development of granulomas and fibrosis will be essential for understanding how schistosome antigens interact with the host environment . We report here the first whole genome microarray analysis of the murine liver during the progression of Schistosoma japonicum egg-induced granuloma formation and hepatic fibrosis . Our results reveal a distinct temporal relationship between the expression of chemokine subsets and the recruitment of cells to the infected liver . Genes up-regulated earlier in the response included T- and B-cell chemoattractants , reflecting the early recruitment of these cells illustrated by flow cytometry . The later phases of the response corresponded with peak recruitment of eosinophils , neutrophils , macrophages and myofibroblasts/hepatic stellate cells ( HSCs ) and the expression of chemokines with activity for these cells including CCL11 ( eotaxin 1 ) , members of the Monocyte-chemoattractant protein family ( CCL7 , CCL8 , CCL12 ) and the Hepatic Stellate Cell/Fibrocyte chemoattractant CXCL1 . Peak expression of macrophage chemoattractants ( CCL6 , CXCL14 ) and markers of alternatively activated macrophages ( e . g . Retnla ) during this later phase provides further evidence of a role for these cells in schistosome-induced pathology . Additionally , we demonstrate that CCL7 immunolocalises to the fibrotic zone of granulomas . Furthermore , striking up-regulation of neutrophil markers and the localisation of neutrophils and the neutrophil chemokine S100A8 to fibrotic areas suggest the involvement of neutrophils in S . japonicum-induced hepatic fibrosis . These results further our understanding of the immunopathogenic and , especially , chemokine signalling pathways that regulate the development of S . japonicum-induced granulomas and fibrosis and may provide correlative insight into the pathogenesis of other chronic inflammatory diseases of the liver where fibrosis is a common feature .
Schistosomiasis , a parasitic disease caused by trematodes of the genus Schistosoma , is a significant cause of human morbidity and mortality . Furthermore , recent reports suggest that the global burden of disease due to schistosomiasis has been significantly underestimated [1] . Chronic infection with S . mansoni or S . japonicum leads to hepatosplenic schistosomiasis , periportal fibrosis , portal hypertension , hepatosplenomegaly , ascites and gastrointestinal bleeding that may lead to death [2] . Murine models of S . mansoni infection indicate that most pathology is attributable to a CD4+ Th2 driven granulomatous response against schistosome eggs and the antigens they secrete ( reviewed in [2] ) . Early studies suggested that the basic immunopathogenic mechanisms associated with granuloma development and fibrosis were similar in both S . mansoni and S . japonicum infections ( e . g . [3] ) , although the severity and features of infection with these two parasites are known to differ in a number of ways . S . japonicum eggs cluster in the host liver where they induce a more severe granulomatous response that is more neutrophilic than those induced by S . mansoni [4] . Furthermore , the immune response of infected mice to purified secreted egg antigen ( SEA ) is of a delayed type hypersensitivity reaction with S . mansoni infection but is of an immediate type hypersensitivity reaction with S . japonicum [5] , suggesting there may be key differences in the pathways leading to granuloma formation and hepatic fibrosis caused by the two species . We used microarray analysis of the mouse liver during infection with a highly pathogenic Chinese mainland field strain of S . japonicum to better define the molecular mechanisms involved in schistosome-induced immunopathology . Progression of disease from the onset of egg laying through to the development of mature granulomas and hepatic fibrosis was associated with temporal expression of genes with distinct biological functions . The contribution of different chemokine subsets to the pathogenesis of schistosomiasis was further defined and suggests that hepatic stellate cell , macrophage and neutrophil chemotaxis are important in the pathogenesis of schistosome-induced hepatic fibrosis .
All work was conducted with the approval of the Queensland Institute of Medical Research Animal Ethics Committee . Four to six week old female C57BL/6 mice were percutaneously infected with 20 S . japonicum cercariae ( Chinese mainland strain , Anhui population ) . Mice were euthanized at 4 ( n = 7 ) , 6 ( n = 7 ) and 7 ( n = 8 ) weeks post infection ( p . i ) and their livers perfused to obtain adult worms . Three additional mice were used as uninfected controls . An identical time-course experiment was performed for flow cytometry ( n = 5 per group ) . The number of adult worm pairs per mouse was recorded as a measure of parasite burden . Eggs per gram of liver were calculated as a measure of hepatic egg burden as described [6] . Briefly , eggs were extracted from a portion of liver of known mass by overnight digestion in potassium hydroxide . Eggs were then resuspended in 1mL of formalin and the number of eggs in three 5µL aliquots were counted and averaged to calculate eggs per gram of liver . Formalin fixed , paraffin embedded liver sections were stained with Haematoxylin and Eosin ( H&E ) , picosirius red for collagen as a measure of fibrosis , α-smooth muscle actin ( SMA ) immunoperoxidase staining for myofibroblasts/Hepatic Stellate Cells ( HSCs ) [7] , Giemsa for eosinophils and Leder stain for neutrophils [8] . Slides were digitised using the Aperio Slide Scanner ( Aperio Technologies , Vista , USA ) . Granuloma volume density , percent collagen staining ( degree of fibrosis ) and percent positive α-SMA staining were quantified by point counting stereology on H&E , picosirius red , and α-SMA stained slides respectively , where myofibroblasts/HSCs were defined as α-SMA positive , spindle shaped cells associated with focal areas of inflammation [7] . Semi-quantitation of eosinophils and neutrophils was performed by determining the mean number of positive-stained cells over 20 fields at high magnification ( cells/hpf ) ( ×400 ) . Immunohistochemistry for S100A8 and CCL7 was performed on paraffin embedded sections using commercially available primary antibodies ( Santa Cruz Biotechnology Inc , Santa Cruz , USA ) and detection kits ( Biocare Medical , Concord , USA ) . Positive staining was quantified using Aperio's Spectrum Plus software positive pixel count algorithm ( Version 8 . 2 . 395 . 1255; Aperio Technologies , Vista , USA ) . Each mouse group was normalised for egg burden by log transformation with outliers excluded on the basis of 95% confidence intervals . Total RNA was extracted from liver tissues using Trizol ( Invitrogen , Carlsbad , USA ) and an RNeasy Mini Kit ( Qiagen Inc , Valencia , USA ) [9] . Total RNA quantity was measured using a Nanodrop-1000 ( Nanodrop Technologies , Wilmington , USA ) and quality was assessed using an Agilent Bioanalyzer ( Agilent Technologies , Foster City , USA ) . Equal amounts of four total RNA samples of the highest quality from each group were pooled for cRNA and cDNA synthesis . cDNA was synthesised using a Quantitect Reverse Transcription kit ( Qiagen Inc . , USA ) . cDNA concentration was measured using a Nanodrop-1000 ( Nanodrop Technologies , Wilmington , USA . ) . Real time PCR was used to validate a subset of microarray data . Genes and primers used for real time PCR were representative of transcripts that were significantly up or down-regulated during microarray analysis and were sourced from the literature [10] , [11] , [12] , [13] , [14] , [15] or designed using Primer 3 software ( http://biotools . umassmed . edu/bioapps/primer3_www . cgi ) ( Table S1 ) . Hypoxanthine phosphoribosyltranferase ( HPRT ) was used as a housekeeping gene . Real time PCR was performed using SYBR Green master mix ( Applied Biosystems , Warrington , UK ) on a Corbett Rotor Gene 6000 ( Corbett Life Sciences , Concorde , Australia ) . Rotor-Gene 6000 Series software ( version 1 . 7 ) and Microsoft Office Excel 2003 were used to analyse the results . Correlation between real-time PCR and microarray data was performed in GraphPad Prism Version 5 . 00 ( GraphPad Software , San Diego , USA ) using Spearman's Rho measure of correlation as described [16] . Leukocytes were isolated from liver tissue as described [17] . Briefly , liver tissue was digested in collagenase D ( Roche Diagnostics , Mannheim , Germany ) ( 1mg/ml ) and DNAse I ( 0 . 5mg/mL , Roche Diagnostics , Mannheim , Germany ) for 45 mins at 37°C . The tissue was then passed through a 70µm cell strainer ( BD Falcon , Bedford , USA ) and washed with phosphate buffered saline supplemented with 2% ( v/v ) Foetal Calf Serum ( FCS ) . The cell pellet was then resuspended in 33% Percoll ( w/v ) and centrifuged at 1700rpm at room temperature to remove hepatocytes and other debris . The resulting leukocyte pellet was then washed in 2% FCS and red blood cells were lysed with Gey's lysis solution . The solution was then underlayed with 2% ( v/v ) FCS and centrifuged at 1300rpm for 5 mins . The resulting cell pellet was resuspended in FACS buffer ( 1% bovine serum albumin ( w/v ) , 0 . 1% sodium azide ( v/v ) in phosphate buffered saline ) and the cells were counted . Leukocytes were stained for specific cell markers by first incubating with antibodies against the Fc-receptorIII ( Monoclonal antibody producing hybridoma; Clone: 24 . G2 ) to block non-specific binding and then with commercially available fluorochrome-conjugated antibodies ( APC-anti-F4/80 , PE-Cy5-anti-CD11b: Biolegend; APC-anti-CD4 , FITC-anti-CD8b , PE-anti-CD19: BD Pharminogen; FITC-CD3 , Miltenyi Biotec , Germany ) . Cell populations were defined as CD4+ T-cells ( CD4+CD3+ ) , CD8+ T-cells ( CD8+CD3+ ) , B-cells ( CD19+ ) and Macrophages ( F4/80+CD11b+ ) . Data were acquired on FACS Calibur Flow Cytometer ( BD Bioscience ) and analysed using FlowJo Software ( Treestar Inc ) and GraphPad Prism , version 5 . 0 ( GraphPad Software , San Diego , USA ) . Changes in parasitological , histological and real time PCR data were assessed by One-Way ANOVA with post hoc Bonferroni testing ( p≤0 . 05 ) . These analyses were performed using the GraphPad Prism Version 5 . 00 ( GraphPad Software , San Diego , USA ) .
Infected mice harboured a mean of 5 worm pairs ( Figure 1a ) . Schistosome eggs were first observed in the liver at 4 weeks p . i and hepatic egg burden increased significantly thereafter ( 1-Way ANOVA , p≤0 . 05 ) ( Figure 1b ) . Granuloma volume increased significantly from 4 weeks p . i , reaching 51% total liver volume at 7 weeks p . i ( 1-Way ANOVA , p≤0 . 01 ) ( Figure 1c ) . Hepatic fibrosis , as measured by collagen staining , in the livers of mice at 4 weeks p . i ( 3% total liver volume ) was not significantly different to that of uninfected controls ( 1% total liver volume ) ( Figure 1d ) . Fibrosis increased from 4–6 weeks post infection but this change was not significant ( 6 weeks: 12% p>0 . 05 1 Way ANOVA ) . The degree of hepatic fibrosis increased further at 7 weeks p . i and was significantly greater than all other time points at 28% total liver volume ( p≤0 . 05 , 1-Way ANOVA ) . Eosinophil numbers increased significantly from 4 weeks p . i . ( p≤0 . 01 ) ( Figure 2a ) . Eosinophils were first observed in small inflammatory infiltrates adjacent to blood vessels and , later , within mature granulomas ( Figure 2b , c ) . Numbers of neutrophils in the liver increased significantly from 6 weeks p . i . ( Figure 2d ) . Neutrophils were first observed in small inflammatory infiltrates ( 6 weeks p . i . ) and , later , in the centre of established granulomas adjacent to schistosome eggs and at the periphery of more fibrotic granulomas ( 7 weeks p . i . ; Figure 2e , f ) . α-SMA staining for myofibroblasts/HSCs was localised to the fibrotic zone of granulomas and was increased significantly compared with uninfected controls at 7 weeks p . i ( 1-Way ANOVA , p≤0 . 001 ) corresponding with increased collagen staining and the development of fibrosis ( Figure 2g–i ) . The expression patterns established by real time PCR correlated well with the microarray data ( Spearman's correlation; r = 0 . 74; p≤0 . 001 ) ( Figure S1 ) . Real-time PCR indicated that expression of IL-4 ( NM_021283 ) peaked at 6 weeks p . i . ( 24-fold relative to uninfected controls , p≤0 . 01 ) and declined thereafter . IL-13 ( NM_008355 ) expression peaked at 6 ( 5-fold , p≤0 . 001 ) and 7 weeks p . i . ( 4-fold , p≤0 . 001 ) during the development of fibrosis . IL-10 ( NM_010548 ) expression reached 5-fold ( p≤0 . 01 ) at 6 weeks and 7-fold ( p≤0 . 01 ) at 7 weeks p . i . Adult worms numbers and hepatic egg burdens were identical in the separate time courses performed for microarray and flow cytometry analyses ( t-test , p>0 . 05 ) ( Figure S2 ) . Analysis of the cellular composition of the liver revealed a significant increase in CD4+ T-cells , CD8+ T-cells and B-cells at 4 weeks p . i , compared with uninfected mice where the numbers of these cells in the liver declined thereafter ( Figure 4a–c ) . The number of macrophages in the liver was significantly increased from 4 weeks p . i . compared with uninfected controls and peaked at 6–7 weeks p . i ( Figure 4d ) . Due to the striking up-regulation of S100A8 and the previous association of CCL7 with fibrosis [20] , immunohistochemistry was used to investigate the distribution of these two chemokines within granulomatous liver tissue . Positive staining for S100A8 was remarkably similar to that for neutrophils . S100A8 positive cells occurred sporadically in the uninfected liver ( Figure 5a ) . Focal clusters of positively stained cells were observed in the liver at 4 weeks p . i . and staining intensity increased significantly thereafter . At 6 and 7 weeks p . i , S100A8 positive cells localised to the inner most zone of established granulomas , where inflammatory cells including neutrophils and macrophages are known to accumulate [4] ( Figure 5b ) , and to the periphery of more mature granulomas adjacent to the fibrotic areas ( Figure 5c ) as confirmed by staining for collagen ( Figure 5d ) . CCL7 was absent from uninfected livers but was present in infected livers from 6 weeks p . i and was significantly elevated above control levels at 7 weeks p . i . Staining occurred predominantly in granulomas at the periphery of the liver ( Figure 5e , f ) and localised to the fibrotic zone of these lesions ( Figure 5g ) resembling the distribution of HSCs ( Figure 5h ) .
We used whole genome microarray analysis and real-time PCR combined with histology and flow cytometry to build a more integrated and global view of the gene signalling pathways and pathological mechanisms induced during S . japonicum infection than has been described previously . Our analyses confirm the development of a Th2 response during S . japonicum-induced granuloma formation and fibrosis , characterised by up-regulation of Th2 associated genes . Notably , we also observed sustained up-regulation of Th1 associated genes including the Th1 cytokine IFN- γ during granuloma development as previously reported in S . japonicum infection [21] . Together this suggests that the localised immune response to S . japonicum eggs in the liver may be of a mixed Th1/Th2 phenotype . Further , the early up-regulation of Th1 associated chemokines [22] , [23] in our study suggests that this arm of the response may be important in the early recruitment of inflammatory cells to the liver and the initiation of the granulomatous response . Widespread down-regulation of multiple components of many metabolic pathways in the livers of S . japonicum infected mice is indicative of a generalised down-regulation of the metabolic functions of the liver . These observations are consistent with those from S . mansoni infections and likely reflects increasing impairment of liver function associated with progressive tissue damage [24] . Genes up-regulated during infection were temporally associated with distinct biological functions . This was especially true for chemokines with activity for distinct cell types which were up-regulated during different phases of the granulofibrotic response ( Figure 6 ) . Up-regulated gene expression in the early phase of S . japonicum-induced granuloma formation ( 4–6 weeks p . i ) was primarily associated with antigen presentation and cell recruitment . Chemokines up-regulated during this phase were predominantly T-cell and B-cell chemoattractants [25] , [26] , [27] , [28] , [29] and their activity was reflected by a concurrent increase in the expression of T- and B-cell makers and the early recruitment of CD4+ T-cells , CD8+ T-cells and B-cells into the liver , consistent with previous studies showing that these cells are required for the development of S . japonicum induced granulomas [30] , [31] . Peak expression of the B-cell chemokine CXCL13 correlated well with the expression of the B-cell maturation marker CD22 but occurred after the influx of CD19+ B-cells into the liver . These results suggest that CXCL13 may not be crucial for the recruitment of B-cells to the liver but may be associated with maturation and/or retention of activated B-cells in the liver . Similar roles for CXCL9 , CXCL13 and CXCL16 in T- and B-cell recruitment have been suggested in other models of liver disease including chronic Hepatitis C , Hepatocellular Carcinoma , Primary Biliary Cirrhosis and Primary Sclerosing Choliangitis [32] , [33] , [34] . We also observed early and sustained up-regulation of the eosinophil chemoattractant CCL24 , which is likely contributing to the ongoing recruitment of eosinophils to the liver as illustrated histologically ( Figure 2 ) . The related chemokine CCL11 was significantly up-regulated later in infection mirroring the accumulation of eosinophils in the liver and correlating well with the expression of eosinophil-associated genes . The late up-regulation of CCL11 suggests a role in the recruitment of eosinophils during the fibrotic response , an observation consistent with the association of this chemokine with fibrosis in chronic liver disease [35] . In accordance with its proposed role in promoting S . mansoni-induced granuloma formation [2] , CCL3 expression peaked during granuloma development . Further , we show a correlation between the expression of CCL3 , several procollagen genes and the profibrotic cytokines TGF-β1 and PDGF- β . This is in agreement with a recent study implicating CCL3 in the development of bleomycin-induced pulmonary fibrosis in the mouse where CCL3 was shown to regulate the recruitment of TGF-β1-producing macrophages and bone-marrow-derived fibrocytes [36] . Together , these results suggest that CCL3 may promote S . japonicum-induced fibrogenesis via a mechanism similar to that proposed for bleomycin-induced pulmonary fibrosis [36] . The related chemokine CCL4 had a similar expression pattern suggesting that it may also play a role in fibrogenesis . Toll-like receptor ( TLR ) signalling was a prominent feature of the mid-late phase of the granulofibrotic response . TLR4 signalling has been shown to be critical to the development of hepatic fibrosis in several experimental models [37] . Further , signalling through TLR4 on quiescent HSCs sensitises these cells to TGF-β1 thereby inducing their activation , chemokine secretion and the chemotaxis of Kupffer cells [38] . The co-incidence of peak TLR expression with that of the pro-fibrogenic cytokines IL-13 and TGF-β1 in our study implies that TLR signalling may be involved in the development of S . japonicum-induced fibrosis . Expression of the chemokines CCL21 , CXCL1 , CCL7 and CCL12 showed a significant correlation with the expression of procollagen genes and resembled the recruitment of HSCs into granulomas . These chemokines have been shown to be chemoactive for HSCs or myofibroblasts and have been implicated in the pathogenesis of a variety of fibrotic diseases [39] , [40] , [41] , [42] . CCL21 , CCL7 and CXCL1 are also known to directly induce the activation and wound healing responses of HSCs or myofibroblasts [20] , [39] , [40] . As well , CCL7 was recently shown to work synergistically with TGF-β1 to induce collagen production in dermal fibroblasts [43] . Furthermore , we showed that CCL7 expression correlates with the expression of fibrosis associated genes and localises to the fibrotic zone of granulomas with a similar distribution to HSCs during S . japonicum infection ( Figure 5e–h ) . The induction of CCL7 and CCL12 during fibrosis is consistent with the role of the related chemokine CCL2 in other liver diseases [44] . CCL2 is a chemoattractant for HSCs [44] and its expression is associated with fibrogenesis in both human cholestatic liver disease and in the bile-duct-ligated rat model of cholestatic liver injury [45] . CCL2 expression is important for the development of granulomas and fibrosis in schistosomiasis mansoni [46] . In contrast , there was no significant change in its expression during S . japonicum infection; instead , CCL21 , CXCL1 , CCL12 and particularly CCL7 , may promote the initiation of S . japonicum induced fibrosis by recruiting fibrogenic effector cells and by directly inducing fibrogenic signalling pathways . Genes with peak expression later in the granulofibrotic response ( 7 weeks p . i ) included many genes associated with fibrosis , such as matrix metalloproteinases ( MMPs ) and tissue inhibitors of matrix metalloproteinases ( TIMPs ) . MMPs and TIMPs play an important role in remodelling of fibrotic tissue and the ratio of MMP:TIMP expression may be a determining factor in the outcome and severity of schistosome induced fibrosis [47] . Up-regulation of MMP-2 , MMP-9 , MMP-13 , TIMP-1 and TIMP-2 appears to be common to both S . japonicum and S . mansoni induced fibrosis [13] , [47] . In contrast , we observed no change in the expression of MMP-8 or MMP-12 , although expression of these genes has been shown to correlate with peak fibrosis during S . mansoni infection [13] , [47] . This is the first report of the up-regulation of MMP-23 and MMP-25 during murine schistosomiasis . These dissimilarities in MMP expression are suggestive of differences in the wound healing response during S . mansoni and S . japonicum infection , which may partly explain the differing degree of fibrosis induced by these two parasites . Chemokines whose peak expression correlated with peak fibrosis and the expression of COL1A1 were predominantly macrophage and neutrophil chemoattractants . CXCL14 , a chemokine which in humans is chemoactive for monocytes [48] , showed peak expression of 10-fold at 7 weeks and therefore may also contribute to the recruitment of monocytes/macrophages to granulomas . This is the first report of the up-regulation of CXCL14 during schistosome infection . Expression of macrophage chemokines corresponded with the up-regulation of macrophage genes and alternatively activated macrophage markers and coincided with the accumulation of F4/80+ macrophages in the liver . Together these results indicate these chemokines contribute to the recruitment of macrophages to the liver during S . japonicum infection and that these macrophages are of an alternatively activated phenotype in accord with the proposed role of these cells in regulating S . mansoni induced inflammation and fibrosis , and other Th2 dominant inflammatory diseases [49] , [50] . Up-regulation of CCL6 , CCL7 and CCL8 has previously been reported for S . mansoni infection [10] , [13] , [22] and so might represent a common mechanism whereby macrophages are recruited into schistosome-induced granulomas . The marked up-regulation of neutrophil chemokines and the number of neutrophils in the liver at 7 weeks p . i . was striking . Neutrophil recruitment has been associated with the development of fibrosis in a number of other chronic liver diseases [51] , [52] . The precise role of these cells in the fibrotic response , however , remains ambiguous . Harty et al [51] reported a role for macrophage mediated neutrophil recruitment in the resolution of fibrosis in a rat model of cholestatic liver disease . In contrast , other studies , including Th1 and Th2 polarised models of S . mansoni infection , have reported an association between increased neutrophil accumulation and the up-regulated expression of CXCL1 with the development of severe disease [52] , [53] , [54] . We localised S100A8 and neutrophils to an area adjacent to the fibrotic zone of mature granulomas ( Figure 5b , c ) , which suggests an accessory role for these cells in S . japonicum induced fibrosis . Whether this role is in promoting or regulating the fibrotic response remains to be determined . In summary , we present the most comprehensive study to date of the transcriptional profile of the schistosome infected liver in the mouse model . It shows that cellular recruitment to granulomatous tissue is tightly regulated by the temporal expression of distinct chemokine subsets and details for the first time the up-regulation of CXCL7 and CXCL14 during schistosome-induced granuloma formation . Additionally , further evidence is provided that there are discrete differences in the cytokine , chemokine and wound healing responses in S . japonicum and S . mansoni infections and indicates that neutrophils may play a significant role in determining the outcome of S . japonicum induced pathology . Furthermore , similarities between the results presented here for schistosomiasis japonica and other chronic inflammatory diseases of the liver suggest that common immunopathogenic pathways regulate the development of hepatic fibrosis in a variety of pathologies .
|
Schistosomiasis , a disease caused by parasitic worms , is a significant cause of illness and death in the developing world . Furthermore , recent reports suggest that the global burden of disease due to schistosomiasis has been significantly underestimated . Schistosomiasis of the liver arises due to inflammation and the deposition of scar tissue around parasite eggs trapped in this organ . In the current study we analysed the gene-expression profile of the mouse liver at several time points following infection with a virulent strain of Schistosoma japonicum to better understand the mechanisms that regulate this process . Progression of disease was associated with increased expression of different groups of genes with distinct biological functions . Specifically , we identified several genes encoding chemical signalling molecules that contribute to different phases of the response by recruiting key cell types to the site of inflammation . This study represents the most comprehensive report to date of the gene expression profile in the liver during schistosomiasis . These results provide further insight into the mechanisms that regulate the development of schistosome-induced inflammation and scarring and will aid in the development of novel treatments to alleviate the burden of disease caused by this parasite .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"immunology/cellular",
"microbiology",
"and",
"pathogenesis",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"immunology/immune",
"response",
"gastroenterology",
"and",
"hepatology",
"infectious",
"diseases/helminth",
"infections",
"microbiology/parasitology",
"pathology/immunology",
"immunology/leukocyte",
"signaling",
"and",
"gene",
"expression",
"pathology/molecular",
"pathology"
] |
2010
|
Temporal Expression of Chemokines Dictates the Hepatic Inflammatory Infiltrate in a Murine Model of Schistosomiasis
|
Schistosomiasis remains a major public health issue , with an estimated 230 million people infected worldwide . Novel tools for early diagnosis and surveillance of schistosomiasis are currently needed . Elevated levels of circulating microRNAs ( miRNAs ) are commonly associated with the initiation and progression of human disease pathology . Hence , serum miRNAs are emerging as promising biomarkers for the diagnosis of a variety of human diseases . This study investigated circulating host miRNAs commonly associated with liver diseases and schistosome parasite-derived miRNAs during the progression of hepatic schistosomiasis japonica in two murine models . Two mouse strains ( C57BL/6 and BALB/c ) were infected with a low dosage of Schistosoma japonicum cercariae . The dynamic patterns of hepatopathology , the serum levels of liver injury-related enzymes and the serum circulating miRNAs ( both host and parasite-derived ) levels were then assessed in the progression of schistosomiasis japonica . For the first time , an inverse correlation between the severity of hepatocyte necrosis and the level of liver fibrosis was revealed during S . japonicum infection in BALB/c , but not in C57BL/6 mice . The inconsistent levels of the host circulating miRNAs , miR-122 , miR-21 and miR-34a in serum were confirmed in the two murine models during infection , which limits their potential value as individual diagnostic biomarkers for schistosomiasis . However , their serum levels in combination may serve as a novel biomarker to mirror the hepatic immune responses induced in the mammalian host during schistosome infection and the degree of hepatopathology . Further , two circulating parasite-specific miRNAs , sja-miR-277 and sja-miR-3479-3p , were shown to have potential as diagnostic markers for schistosomiasis japonica . We provide the first evidence for the potential of utilizing circulating host miRNAs to indicate different immune responses and the severity of hepatopathology outcomes induced in two murine strains infected with S . japonicum . This study also establishes a basis for the early and cell-free diagnosis of schistosomiasis by targeting circulating schistosome parasite-derived miRNAs .
Schistosomiasis is a chronic debilitating parasitic disease of humans . Caused by members of the genus Schistosoma , it afflicts more than 200 million individuals worldwide , representing a major health and economic burden in tropical and developing nations [1] . The pathology of chronic Schistosoma japonicum or Schistosoma mansoni infection , in its severe form , results in hepatosplenic schistosomiasis , with clinical symptoms of granuloma formation , periportal fibrosis , portal hypertension , hepatosplenomegaly , ascites , and the formation of vascular shunts [1] . The granuloma formation is characterised by a focussed accumulation of a group of specific immune cells around the schistosome eggs , followed by a fibrosing lesion , which forms as a zone of collagen at its periphery [2 , 3] . Both features , while beneficial in limiting and neutralising the toxicity of secreted egg antigens ( SEA ) released from parasite eggs , also cause reversible hepatic damage , indicating that the immune-cellular response is both friend and foe to the schistosome-infected host [4 , 5] . The general epidemiological situation of schistosomiasis in China has changed due to long-term extensive and integrated control efforts [6 , 7] . A number of endemic areas are close to transmission interruption and thus improved diagnostic tools are urgently needed for the surveillance of control efforts to ensure that the elimination of schistosomiasis can be achieved [7] . Currently , there are four major methods available for the diagnosis of schistosomiasis: parasitological detection ( PD; mainly by the Kato-Katz method ) , antibody-detection ( AbD ) , antigen-detection ( AgD ) , and the detection of circulating schistosome nucleic acids ( CNAD ) by PCR . The former three procedures have disadvantages in that they exhibit low sensitivity ( e . g . PD ) , cross-reactivity with other helminth infections or cannot distinguish between active and past infections ( e . g . AbD and AgD ) , with the latter limitation particularly important in endemic areas . The requirement for the detection of schistosome ova in faeces and/or SEA-specific serum antibodies limits such methods for early diagnosis before patency . These inadequacies demonstrate the current limitations in monitoring the progress of schistosomiasis control , especially in areas with low schistosome prevalence or low levels of transmission [8] . MicroRNAs ( miRNAs ) are a class of small non-coding RNAs approximately 22 nucleotides in length , which can be detected in a wide range of body fluids , including blood plasma/serum [9 , 10] . The high stability of miRNAs in biofluids has been mainly attributed to two mechanisms: ( 1 ) formation of a protein-miRNA complex with argonaute proteins or high-density lipo-proteins , and ( 2 ) their incorporation into exosomes [11] . MiRNAs have been increasingly regarded as promising targets for the next generation of diagnostic biomarkers as the strong correlation between the status/progression of various diseases and the dysregulated profile of miRNAs has been confirmed . The potential for detecting circulating miRNAs as biomarkers for various cancers , viral infections , as well as drug-induced liver injury , has been widely reported [9 , 12–17] . Previously , a panel of host miRNAs was shown to be dysregulated in murine hepatic tissue with the progression of schistosomiasis , highlighting the fact that miRNAs may play a variety of regulatory roles in the immunological responses that occur during the development of hepatopathology [18 , 19] . The altered expression profile of hepatic miRNAs during schistosomal infection differed from that of other liver diseases [20] , indicating that schistosome egg-induced hepatic immunopathology is a unique type of chronic liver disease , distinguishable from many other types of liver disease . Though the diagnostic and therapeutic potential of parasite-derived miRNAs have been discussed [21] , the area is still in the early stages of infancy . Nevertheless , five schistosome-specific miRNAs were identified in the plasma of rabbits infected with S . japonicum using a deep sequencing method and one of them , sja-miR-3479-3p , showed diagnostic potential for S . japonicum infection [22] , although further confirmation in other animal models and in patients is required . Further , the presence of three parasite-derived miRNAs in serum discriminated patients infected with S . mansoni from normal individuals [18] and circulating parasite-derived miRNAs have been found in the plasma or serum of dogs with a filarial worm infection [23] . Regarding host circulating miRNAs , inconsistent results have been observed in different mouse models of schistosomiasis . For example , the level of liver-specific miR-122 was elevated in the serum of BALB/c mice after S . japonicum infection [24] , while it did not change in the serum of C57BL/6 mice between 4–12 weeks post-S . mansoni infection [18] . Since these two mouse strains induced differential pathological outcomes , including the severity of hepatic granulomatous pathology and fibrosis at some particular time points post-schistosome infection [25 , 26] , these observations led us to suspect that hepatopathology progression of schistosomiasis may significantly affect the abundance of host miRNAs in serum . Despite these recent studies , there is generally limited information about the diagnostic value of circulating miRNAs in parasitic diseases and their associated pathologies . We hypothesise that both host- and schistosome parasite-derived miRNAs in serum may present a dysregulated profile during the progression of hepatic schistosomiasis , thereby providing promising targets for an early and cell- diagnosis for the disease . In addition , circulating miRNAs of host origin may provide highly sensitive molecular signatures for the assessment of hepatopathology severity induced by schistosome eggs . Two mouse strains , C57BL/6 and BALB/c mice , were employed to verify our hypothesis .
All work was conducted with the approval of the QIMR Berghofer Medical Research Institute Animal Ethics Committee ( Ethics Approval: Project P288 ) . Animal studies were conducted according to the Australian Code for the Care and Use of Animals for Scientific Purposes ( 8th edition ) and the protocols approved by the QIMR Berghofer Medical Research Institute Animal Ethics Committee . Eight-week-old female C57BL/6 and BALB/c mice were percutaneously infected with 14 S . japonicum cercariae ( Chinese mainland strain , Anhui population ) . Mice were euthanized at 4 , 6 , 7 , 9 , 11 ( both mouse strains ) and 13 ( C57BL/6 only ) weeks post infection ( p . i . ) . Since BALB/c mice are more susceptible to S . japonicum infection , the experiment with this strain lasted for 11 weeks p . i . as prolonging the time of infection to 13 weeks would have resulted in premature death of many of the animals due to the resulting egg-induced pathology . Ten naive mice were used as controls for each mouse strain . Each experimental group comprised 10 mice at time points 4 , 6 and 7 weeks p . i . , and 12 mice were used at 9 , 11 and 13 weeks p . i . . The liver and blood samples ( ~1 mL ) were collected by cardiac puncture at each time point . Eggs per gram of liver were calculated as a measure of hepatic egg burden and general infection level , as described [25] . Briefly , eggs were extracted from a portion of liver of known mass by overnight digestion with 5% ( w/v ) potassium hydroxide . After centrifugation , eggs were then resuspended in 2 mL of 4% ( v/v ) formalin and the number of eggs in three 10 μL aliquots counted and averaged to calculate the mean eggs per gram of liver ( S1 Table ) . Blood samples were allowed to stand at room temperature for 2 h and then centrifuged at 4 , 000 rpm for 10 min at 4°C , followed by another centrifugation for 10 min at 10 , 000 rpm at 4°C . The supernatants were retained and stored at -80°C . Haemolysed samples were excluded from further analysis ( S1 Table ) . For each mouse , RNA was extracted from 100 μL of serum using the miRNeasy mini kit ( Qiagen , Hilden , Germany ) according to the manufacturer’s protocol . Non-parasitic miRNA ( 3 . 2 fmoles ) , ath-miR-159a , 5′-UUUGGAUUGAAGGGAGCUCUA-3′ ( IDT , Coralville , IA ) was spiked to each denatured sample to normalize the technical variability of the serum RNA extraction . For each sample , the final RNA product was eluted into 30 μL nuclease-free water and stored at -80°C prior to further analysis . In some experimental groups , blood samples from unisexually infected or uninfected mice without showing any signs of hepatopathology were excluded from further analysis ( S1 Table ) . Polyadenylation and RT reactions were performed with S-Poly ( T ) method with minor modifications to a published protocol [27] . Total RNA was polyadenylated with a Poly ( A ) polymerase tailing kit ( Epicentre Biotechnologies , Madison , WI ) and the first-strand cDNA was synthesized using a TaqMan microRNA reverse transcription kit ( Life Technologies , Carlsbad , CA ) in a 10 μL RT reaction: 2 . 53 μL H2O , 1 μL 10 × PAP buffer , 0 . 1 μL ATP ( 10 mM ) , 0 . 5 μL miRNA-specific primer pool ( 50 nM for each primer ) , 0 . 04 μL dNTPs ( 25 mM each ) , 0 . 13 μL RNase inhibitor , 0 . 2 μL Poly ( A ) polymerase , 0 . 5 μL MultiScribe MuLV and 5 μL RNA . Reverse transcription ( RT ) reactions were conducted using a Veriti 96-well thermal cycler ( ABI ) under the following conditions: 42°C for 60 min , 95°C for 5 min . RT products were stored undiluted at -20°C prior to the further qRT-PCR reactions . A list of all the primers used in this study is presented in S2 Table . The efficiency of PCR amplification for each primer pair was evaluated by creating a standard curve plot for 10-fold serial dilutions of PCR product ( S2 Table ) . Quantification of serum miRNAs was performed according to qPCR protocols described previously [28] . Briefly , the 10 μL PCR reaction contained 3 μL H2O , 0 . 5 μL of RT products ( 2 . 5× dilution ) , 0 . 5 μL forward primer , 0 . 5 μL universal reverse primer ( final conc: 0 . 2 μM ) , 0 . 5 μL universal double-quenched probe ( 56-FAM/CAGAGCCAC/ZEN/CTGGGCAATTT/3IABkFQ , final conc: 0 . 25 μM ) ( IDT ) and 5 μL TaqMan Universal Master Mix II ( Life Technologies ) . Amplification was performed on an Applied Biosystems Viia 7 thermal cycler ( Applied Biosystems ) with the cycling conditions: pre-denaturation at 95°C for 10 min , followed by 40 cycles: 95°C for 15 sec , and 60°C for 30 sec . For detecting parasite-derived miRNAs , 50 cycles were performed , and the maximum cycle value of 42 was set as background for the purpose of calculating signal over noise . Spiked-in ath-miR-159a was used as the normalized internal control , and the fold change was calculated by the 2-ΔΔCt method [29] . A comparative analysis was carried out to highlight any concordance with respect to host miRNAs between the two mouse strains based on the log base2-transformed qPCR data . The sequences of the primers used are listed in S2 Table . The PCR products were further examined by 15% TBE-PAGE ( S1 Fig ) . Three technical replicates were performed for each sample and repeated PCR assays were carried out for detection of each miRNA ( S2 Fig ) . A biological replicate was carried out with serum samples from BALB/c mice at 4 and 9 weeks post-infection ( S3 Fig ) . The median lobe from each mouse liver was used for histological assessment . Formalin-fixed , paraffin embedded liver sections were stained with Haematoxylin and Eosin ( H&E ) as a measure of granuloma and necrosis , picosirius red for collagen as a measure of fibrosis , alpha-smooth muscle actin ( α-SMA ) and immunoperoxidase staining for myofibroblasts/Hepatic Stellate Cells ( HSCs ) . Slides were digitised using the Aperio Slide Scanner ( Aperio Technologies , Vista , USA ) . Granuloma volume density and percent of hepatic necrosis , percent of collagen staining ( degree of fibrosis ) and percent of positive α-SMA staining were quantified with an Aperio ImageScope V10 . 2 . 1 with H&E , picosirius red , and α-SMA stained slides , respectively; myofibroblasts/HSCs were defined as α-SMA positive , spindle-shaped cells associated with focal areas of inflammation [25] . Serum alanine transaminase ( ALT ) and aspartate transaminase ( AST ) levels were measured with the ALT and AST colour endpoint assay kits ( Bioo Scientific , Austin , TX ) , respectively , according to the manufacturer’s instructions . All results are reported as means ± SEM ( standard error of the mean ) . For analysis of the serum levels of host miRNAs as well as that of ALT and AST levels during the infection course , one-way ANOVA followed by Holm-Sidak multiple comparison was used . For analysis of relative serum abundance of host miRNAs , hepatic egg burden and histology , two-way ANOVA followed by Holm-Sidak multiple comparisons were used to compare statistical differences between the two mouse strains . For analysis of the parasite-derived miRNAs in serum , the Man-Whitney test was used and p-values of <0 . 05 were considered statistically significant . Associations were measured using Spearman’s Rho correlation in GraphPad Prism Version 6 . 00 for windows .
Two mouse strains , C57BL/6 and BALB/c , were employed as schistosomiasis japonica disease models to detect in serum , four host circulating miRNAs , miR-122 , miR-21 , miR-20a and miR-34a , all of which have been suggested to be correlated with different types of liver disease progression [30–33] . In C57BL/6 mice , the serum concentrations of miR-122 , miR-20a and miR-34a did not change at any time point post infection , but the level of serum miR-21 was increased at 6 ( 1-Way ANOVA , P<0 . 01 ) ( Fig 1A ) . In contrast , apart from miR-20a , the serum levels of the three other host miRNAs were significantly elevated in BALB/c mice by 6 ( miR-122 ) or 7 ( miR-21 and miR-34a ) weeks p . i . and thereafter ( Fig 1B and S3 Fig ) . Similar results were observed by He et al . , who found that the levels of miR-122 and miR-34a were significantly elevated in the serum of BALB/c mice at 72 days post-S . japonicum infection [24] . There was a tendency , albeit not statistically significantly , for relatively high expression of miRNA-20a in the serum of infected BALB/c mice at 7 weeks p . i . and thereafter . This may have been due to the existence of multiple relatively high expressed miRNA-20a isomiRs and the design of the RT-primer against this miRNA , since we only designed one RT-primer against one form of miRNA-20a isomiRs . Thus , the low efficiency of reverse transcription of other miRNA-20a isomiRs might have impaired the sensitivity of detecting this miRNA to some degree . Another explanation is that the expression of miRNA-20a may be down-regulated in the necrotic hepatocytes . Accordingly , we carried out comparative analysis of the relative abundance of these miRNAs in the two mouse strains . With miR-122 and miR-34a , there was a significant difference between the two mouse strains at 4–11 weeks p . i . , while for miR-21 and miR-20a , significant difference was only observed at 7 weeks p . i . ( Fig 1C ) . In light of these differing results between the two strains , we carried out further parasitological , histological and serum chemistry analyses . There were no significant differences in hepatic egg burden between the two mouse strains at any time-point ( 2-Way ANOVA , P>0 . 05 ) ( Fig 2A ) , indicating that differences in hepatic egg burden did not cause the differential serum levels of miR-122 , miR-21 and miR-34a observed in the two mouse strains during S . japonicum infection . Granuloma area was significantly greater in infected C57BL/6 mice compared with BALB/c mice at 9 and 11 weeks p . i . . In C57BL/6 mice , the granuloma area represented 38 . 8% and 37 . 5% of the total liver area at 9 and 11 weeks p . i . , respectively , whereas the granuloma area in BALB/c mice was 19 . 9% and 17 . 9% of the total liver area at the corresponding time points , respectively ( 2-Way ANOVA , P<0 . 0001 ) ( Fig 2B ) . These differences are likely due to the hepatic granuloma area being continually increased in size in C57BL/6 mice between 7~9 weeks p . i . , while it had plateaued by this time in BALB/c mice . Although not statistically significant , granuloma area also showed a tendency towards increased size in C57BL/6 mice compared with BALB/c mice at 6 weeks p . i . ( Fig 3A and 3B ) . Hepatic fibrosis was induced more rapidly in C57BL/6 mice compared with BALB/c mice . This was reflected by significantly greater collagen deposition in C57BL/6 mice at 6 weeks p . i . , where collagen represented 17 . 1% of the total liver area , compared with 6 . 2% in BALB/c mice ( 2-Way ANOVA , P<0 . 01 ) ( Figs 2C , 3C and 3D ) . However , at 7 weeks p . i . there was no significant difference in hepatic fibrosis between the two mouse strains ( 2-Way ANOVA , P>0 . 05 ) . Similar to the granuloma area , hepatic fibrosis in C57BL/6 mice was significantly more intensive than those in BALB/c mice at 9 and 11 weeks p . i . ( 2-Way ANOVA , P<0 . 0001 ) . The differential activation level of myofibroblasts/Hepatic Stellate Cells ( HSCs ) in the two mouse strains showed a similar pattern with hepatic fibrosis ( Figs 2D , 3E and 3F ) . In contrast , H&E staining indicated that hepatic necrosis was significantly more pronounced in BALB/c mice than in C57BL/6 mice at 6 and 11 weeks p . i . ( 2-Way ANOVA , P<0 . 001 and P<0 . 05 , respectively ) ( Fig 2E ) . It is noteworthy that , on average , the necrosis area represented 0 . 73% of the total liver area in C57BL/6 mice , whereas it reached 2 . 39% in BALB/c mice at 6 weeks p . i . ( Figs 2E , 3A and 3B ) . Followed the development of granulomas and liver fibrosis , the severity of hepatic necrosis was alleviated in BALB/c mice and no significant differences were observed between the two mouse strains at 7 and 9 weeks p . i . . Based on the analysis of data from 6–13 weeks p . i . , a significant inverse correlation was observed between the level of hepatic necrosis and the degree of granuloma formation in C57BL/6 mice ( r = -0 . 4393 , P = 0 . 0073 ) , but not in BALB/c mice ( Fig 2F ) . More importantly , significant inverse correlations were observed between the level of hepatic necrosis and fibrosis in both mouse strains ( C57BL/6 , r = -0 . 4908 , P = 0 . 0024; BALB/c , r = -0 . 3484 , P = 0 . 0373 ) ( Fig 2G ) . These observations led us to hypothesise that the up-regulation of serum miR-122 , miR-21 and miR-34a levels in BALB/c mice during infection may be mainly due to the massive release of these miRNAs from necrotic hepatocytes . In order to test this hypothesis , we further examined the dynamic changes in serum levels of the liver injury-related enzymes , aspartate aminotransferase ( AST ) and alanine aminotransferase ( ALT ) . Similar to the temporal alteration of the serum level of miR-21 in C57BL/6 mice , the serum AST level was only significantly elevated at 6 weeks p . i . ( 1-Way ANOVA , P<0 . 05 ) ( Fig 4A ) , whereas the serum ALT level was significantly up-regulated at 6 , 7 and 11 weeks p . i . ( 1-Way ANOVA , P<0 . 001 , P<0 . 05 , and P<0 . 01 , respectively ) . In BALB/c mice , both serum AST and ALT levels were significantly elevated at 6 weeks p . i . and thereafter ( 1-Way ANOVA , P<0 . 0001 , except for AST level at 11 weeks p . i . , P<0 . 001 ) . Further analysis revealed that the serum levels of all four miRNAs were significantly correlated with ALT and AST levels in BALB/c mice ( Fig 4B and 4C ) , whereas only the serum level of miR-21 showed a significant correlation with the serum levels of these liver enzymes in C57BL/6 mice ( Fig 4B and 4C ) . Moreover , as shown in Fig 4D , the serum levels of all four miRNAs were significantly positively correlated with the severity of liver necrosis in BALB/c , but not in C57BL/6 mice , an observation which provided additional direct evidence in support of our hypothesis . The reason why serum miRNA levels did not correlate with the severity of liver necrosis in C57BL/6 mice is largely because the degree of hepatic necrosis in this strain is not as pronounced as in BALB/c mice during the course of S . japonicum infection . Further , the abundance of serum miRNAs in C57BL/6 mice are at baseline levels post-infection . Among these four miRNAs , the serum concentration of miR-122 showed the strongest association with the serum levels of liver injury-related enzymes and the severity of hepatic necrosis in BALB/c mice during S . japonicum infection , and this was followed by miR-21 ( Fig 4B–4D ) . In addition , the abundance of serum miR-34a showed the strongest association with the degree of liver fibrosis in BALB/c mice during schistosomiasis progression , followed by miR-122 and miR-21 ( Fig 4E ) . Using a deep sequencing method , Cheng et al . identified the presence of five schistosome-specific miRNAs ( sja-bantam , sja-miR-3479-3p , sja-miR-10-5p , sja-miR-3096 and sja-miR-8185 ) in the plasma of S . japonicum-infected rabbits . Three of these ( sja-bantam , sja-miR-3479-3p , sja-miR-10-5p ) were further detected in the plasma of S . japonicum-infected mice by stem-loop RT-PCR analysis [22] . More recently , it was shown that S . mansoni-derived miRNAs ( miR-277 , miR-3479-3p and bantam ) in serum could discriminate infected from uninfected individuals [18] . We thus carried out RT-qPCR analysis to determine the dynamic serum levels of five parasite-derived miRNAs ( sja-bantam , sja-miR-3479-3p , sja-miR-3096 , sja-miR-8185 and sja-miR-277 ) during S . japonicum infection . Sja-miR-10-5p was excluded from analysis due to its high sequence homology with mammalian host orthologs . Only two parasite-derived miRNAs ( sja-miR-277 and sja-miR-3479-3p ) could be reliably detected in serum specimens from both mouse strains ( Figs 5A–5D and S3 Fig ) . However , one intriguing feature was the time phase when the serum levels of these two miRNAs started to significantly alter differed in the two models . In BALB/c mice , sja-miR-277 and sja-miR-3479-3p showed a statistically significant signal over noise as early as 4 and 6 weeks p . i . ( Fig 5C and 5D ) , respectively , while in C57BL/6 mice , these signals were delayed to 6 and 9 weeks p . i . ( Fig 5A and 5B ) , respectively . Moreover , both the serum levels of sja-miR-277 and sja-miR-3479-3p were significantly correlated with hepatic egg burdens ( Fig 5E and 5F ) and the degree of liver fibrosis ( Fig 5G and 5H ) in the two mouse strains . However , the serum level of sja-miR-277 showed a stronger correlation with liver fibrosis intensity than that of sja-miR-3479-3p . All key results obtained with the two mouse strains are summarized in Table 1 .
The recent discovery of the extreme stability of circulating miRNA in body fluids and the fact their dysregulated profiles are associated with disease progression are characteristic of a wide variety of diseases and syndromes . These features of miRNAs have trigged widespread interest in their potential as biomarkers for diagnosis and pathologic status of chronic and infectious diseases . This study aimed to investigate whether several circulating host miRNAs commonly associated with liver diseases are dysregulated in murine schistosomiasis japonica and whether S . japonicum-derived miRNAs could be detected in serum specimens from two mouse strains during disease progression , thus determining their potential value as biomarkers for evaluation of the severity of hepatopathology caused by schistosome eggs and the detection of S . japonicum infection , respectively . The advantage of using two mouse strains in this study was the capacity to observe the considerably different dysregulation of circulating host miRNAs , miR-122 , miR-21 and miR-34a , in the sera of C57BL/6 and BALB/c mice during S . japonicum infection . We also assessed the fibrogenic granulomatous response induced by schistosome eggs which leads to the hepatopathology characteristic of schistosomiasis , an area of considerable interest [5 , 34–37] . In contrast , hepatic necrosis is another type of hepatopathology caused by the toxicity due to schistosome eggs , an area which has received much less attention . The key differences in hepatopathology observed between the two mouse strains examined here is centrally linked to the delayed development of fibrosis , with the level of hepatic necrosis in BALB/c mice markedly more extensive than in C57BL/6 mice at 6 weeks p . i . . The different degree of leakage of egg antigens into the adjacent liver tissue at 6 weeks p . i . in the two mouse strains may contribute partially to the intensity of fibrosis latter as soluble egg antigen ( SEA ) of schistosomes has been shown to suppress the activation and facilitate apoptosis of HSCs [37 , 38] . For the first time , we have shown an inverse correlation between the severity of hepatocyte necrosis and the level of liver fibrosis in both mouse models , which further supports the protective role of fibrosis in restricting the SEA within focal areas of chronic inflammation , thus reduce the hepatotoxic effects caused by the eggs trapped in the liver tissue [39] . Also this may explain the differential serum levels of hepatocellular enzymes , as well as different abundances in some serum host miRNAs observed between the two mouse strains . Thus , as summarized in Table 1 , the differential levels of miR-122 , miR-21 and miR-34a in host sera are mainly the result of hepatopathology caused by the different types of immune response induced in C57BL/6 and BALB/c mice after S . japonicum infection , especially following the onset of egg deposition . These three host circulating miRNAs may , as a panel , serve as indicative biomarkers for the severity of hepatopathology outcomes , particularly regarding hepatocytes damage or necrosis , in patients having similar worm burdens . However , further studies with clinical samples are needed to justify this suggestion . It is well known that circulating miRNAs are derived either via passive release of cellular contents from tissue damage or via active secretion of microvesicles/exosomes from cells [40] . Here , significant correlations between the levels of serum miRNAs ( miR-122 and miR-21 ) and liver enzymes indicate that the passive release from injured tissues may represent a key mechanism for the observed increased levels of these miRNAs . MiR-122 is the predominant liver-specific miRNA , constituting about 70% of the total miRNA population in normal liver tissue [41] . This may explain why the significant alteration in miR-122 serum levels could be sensitively detected in BALB/c mice as early as 6 weeks p . i . , at the same time when hepatic necrosis is evident . There are consistent observations that miR-122 serum levels are elevated in a number of liver diseases with different etiologies , suggesting that this miRNA may act as a clear biomarker of general liver injury [11 , 42] . Furthermore , plasma miR-122 has been shown to have a better performance than ALTs in the detection of liver injury [43 , 44] . The serum miR-20a and miR-34a levels showed a significant but weaker correlation with the serum AST and ALT levels than those of miR-122 and miR-21 in BALB/c mice . These data indicate that other tissues , such as the spleen and/or intestine , which also retain schistosome eggs , may contribute to the serum levels of miR-20a and miR-34a , both are multi-tissue expressed miRNAs [45 , 46] . It would be useful to investigate the pathology of other injured organs during schistosomiasis for the complete recognition of potential sources of these new biomarkers . In BALB/c mice , the elevated serum miR-122 and miR-21 levels showed a much stronger correlation with hepatocellular enzymes than with the level of hepatic necrosis . This can be explained by the fact that hepatic necrosis peaked at 6 weeks p . i . in this strain and dramatically decreased thereafter , while the serum levels of miR-122 and miR-21 reached a plateau after 7 weeks p . i . , due to accumulation of these miRNAs , which are extremely stable in body fluids [9 , 47] . Meanwhile , the degree of hepatic granuloma and fibrosis also stabilized after 7 weeks p . i . in BALB/c mice , which resulted in significant positive correlations between the serum miR-122 and miR-21 levels and hepatic fibrosis severity . However , no positive correlation in the serum levels of these four miRNAs and the degree of liver fibrosis was observed in C57BL/6 mice . Hence , it could be misinterpreted that the elevation in serum levels of miR-122 and miR-21 was caused by the fibrogenic granulomatous responses , rather than actually being caused by necrosis , if only the results from BALB/c mice were considered . Previously , He et al . showed that circulating miR-223 could serve as a potential novel biomarker for the detection of S . japonicum infection [24] . However , miR-21 , miR-122 and miR-223 were also shown elevated in the serum of patients with HCC ( hepatic cellular carcinoma ) and chronic hepatitis and these miRNAs were suggested as novel biomarkers for liver injury but not specifically for HCC [32] , thereby providing support that the elevation of serum miRNA-223 level might also be caused by liver necrosis due to S . japonicum infection . Three circulating S . mansoni-derived miRNAs , sma-miR-277 , sma-miR-3479-3p and sma-bantam , have been shown to have potent diagnostic value in detecting S . mansoni infection [18] , but in the current study , only two orthologs , sja-miR-277 and sja-miR-3479-3p , could be reliably detected in the sera of the two mouse strains infected with S . japonicum . This may be due to two reasons: ( 1 ) mice were challenged with a low dose of cercariae compared with the previous study [18]; ( 2 ) the serum level of sja-bantam was comparatively lower than that of sja-miR-277 and sja-miR-3479-3p . We detected significant increases in the serum level of sja-miR-277 in BALB/c and C57BL/6 mice at 4 and 6 weeks p . i . , respectively , earlier than with S . mansoni—8 weeks p . i . - , although a higher challenge was given in their study . This could have a technical explanation that , unlike Hoy et al . , who employed the miScript system to perform the reverse transcription reactions [18] , here , we employed the S-Poly ( T ) method , which has been shown to improve both the specificity and sensitivity of the PCR reaction [27] . As miRNAs usually have different isoforms , known as isomiRs [48 , 49] , the design of multiple RT primers against different isomiRs originated from one particular miRNA may further improve the sensitivity of miRNA detection when using the S-Poly ( T ) method . We found that schistosome egg-induced pathology may also impact on the detection of parasite-derived miRNAs in serum . This was reflected by the differential time phase for detecting sja-miR-277 and sja-miR-3479-3p in serum . Hepatic granulomas and more importantly , fibrosis , tightly enclose most schistosome eggs trapped in the liver , so as to limit the release of the parasite-derived hepatotoxic proteins and miRNAs from eggs into the surrounding tissue . This may have delayed the detection of these miRNAs in the serum of C57BL/6 mice , when compared with BALB/c mice . The significant correlation between the serum level of these parasite miRNAs and the hepatic egg burden indicates that eggs might serve as an important source for these miRNAs , since a single adult worm pair of S . japonicum can release an estimated number of 3 , 000 eggs per day . It is also notable that sja-miR-277 and sja-miR-3479-3p are not the most highly expressed miRNAs in either adult worms or eggs , but are observed at intermediate levels [50] . The most highly expressed miRNAs in the parasite , such as sja-miR-71 , sja-miR-71b-5p and sja-miR-1 [50] , are undetectable in the serum/plasma of animal hosts infected with schistosomes [18 , 22] . This observation suggests that schistosome parasite-derived miRNAs , which may be expressed in a cell- or tissue-specific pattern , are selectively released by adult worms and eggs . Exosomes display significantly different selective enrichment of specific extra-cellular miRNAs compared to those from their source cells [51] . However , it does not exclude the fact that adult worms secrete miRNAs into the circulating blood stream , since we have also detected sja-miR-277 and sja-miR-3479-3p in the serum of mice infected with unisexual male worm ( s ) ( S4 Fig ) . Thus , adult worms and/or eggs contribute to the origin of these parasite-derived miRNAs in serum , which may represent potential markers for the early diagnosis of schistosomiasis . In summary , inconsistent serum levels of host miR-122 , miR-21 and miR-34a in different murine models during infection may impair their value as diagnostic biomarkers for schistosomiasis . However , the serum levels of these miRNAs as a panel may correlate with the hepatic immune responses of a schistosome-infected individual , and they may serve as novel biomarkers to indicate the degree of hepatopathology caused by schistosomiasis . The circulating parasite-specific miRNAs , sja-miR-277 and sja-miR-3479-3p , have potential to be diagnostic markers for schistosomiasis japonica , but the sensitivity for early detection of these miRNAs may not only depend on the parasite load but may also be affected by the host pathology induced by schistosome eggs .
|
Schistosomiasis japonica remains a public health problem in Southeast Asia . In China , the number of infected individuals has been reduced considerably due to long-term control efforts over the past 50 years , but still 60 million people are at the risk of the disease . Development of novel tools for early diagnosis and surveillance of schistosomiasis are urgently needed . Circulating microRNAs are increasingly regarded as promising targets for the next generation of diagnostic biomarkers . This study systematically compared both host- and schistosome parasite-derived circulating miRNAs associated with Schistosoma japonicum infection in two murine models . For the first time , we revealed that host circulating miRNAs dysregulated after S . japonicum infection , are mouse strain-dependent , along with different pathological responses . Three host circulating miRNAs , miR-122 , miR-21 and miR-34a , may , as a panel , serve as indicative biomarkers for hepatopathology progressions . We also confirmed previous reports of the value of parasite-specific miRNAs ( sja-miR-277 and sja-miR-3479-3p ) in serum as potential biomarkers for the diagnosis of schistosomiasis japonica . This study establishes a basis for using miRNAs as supplemental biomarkers for the early and cell free diagnosis of schistosomiasis .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Circulating miRNAs: Potential Novel Biomarkers for Hepatopathology Progression and Diagnosis of Schistosomiasis Japonica in Two Murine Models
|
The syndrome of fever is a commonly presenting complaint among persons seeking healthcare in low-resource areas , yet the public health community has not approached fever in a comprehensive manner . In many areas , malaria is over-diagnosed , and patients without malaria have poor outcomes . We prospectively studied a cohort of 870 pediatric and adult febrile admissions to two hospitals in northern Tanzania over the period of one year using conventional standard diagnostic tests to establish fever etiology . Malaria was the clinical diagnosis for 528 ( 60 . 7% ) , but was the actual cause of fever in only 14 ( 1 . 6% ) . By contrast , bacterial , mycobacterial , and fungal bloodstream infections accounted for 85 ( 9 . 8% ) , 14 ( 1 . 6% ) , and 25 ( 2 . 9% ) febrile admissions , respectively . Acute bacterial zoonoses were identified among 118 ( 26 . 2% ) of febrile admissions; 16 ( 13 . 6% ) had brucellosis , 40 ( 33 . 9% ) leptospirosis , 24 ( 20 . 3% ) had Q fever , 36 ( 30 . 5% ) had spotted fever group rickettsioses , and 2 ( 1 . 8% ) had typhus group rickettsioses . In addition , 55 ( 7 . 9% ) participants had a confirmed acute arbovirus infection , all due to chikungunya . No patient had a bacterial zoonosis or an arbovirus infection included in the admission differential diagnosis . Malaria was uncommon and over-diagnosed , whereas invasive infections were underappreciated . Bacterial zoonoses and arbovirus infections were highly prevalent yet overlooked . An integrated approach to the syndrome of fever in resource-limited areas is needed to improve patient outcomes and to rationally target disease control efforts .
Fever without a localized cause is one of the most common presenting complaints among persons seeking healthcare in many low- and middle-income countries [1] , [2] . However , unlike the syndromes of pneumonia and diarrhea that feature in global disease burden estimates and have well coordinated programs integrating efforts across the range of responsible pathogens to avert morbidity and mortality , there has been a lack of a coordinated approach for febrile illness . While illness and death due to some specific infections causing fever , such as malaria [3] and increasingly bacterial sepsis are well quantified [4]–[6] , others such as a range of zoonoses and viral infections are uncounted and consequently may be underappreciated . The various etiologies of febrile illnesses are difficult to distinguish from one another clinically [7] , [8] . As clinical laboratory services are often limited in areas where febrile conditions are particularly common [9] , [10] , clinicians may have few diagnostic tools to establish an etiologic diagnosis . Therefore , clinical management is often driven by syndrome-based guidelines employing empiric treatment [11]–[13] . In the absence of systematically collected data on fever etiology , considerable mismatch between clinical diagnosis , clinical management , and actual etiology may occur resulting in poor patient outcomes [14] . It is increasingly recognized that malaria is over-diagnosed in many areas [14] , [15] . To address this problem , the World Health Organization ( WHO ) malaria treatment guidelines moved away from clinical diagnosis of malaria to treatment based on the results of a malaria diagnostic test such as a blood smear or a malaria rapid diagnostic test . With more widespread availability of diagnostic tests to exclude malaria and apparent declines in malaria worldwide [3] , clinicians in resource-limited areas are faced with a growing proportion of febrile patients who do not have malaria and few tools to guide subsequent management . We sought to describe comprehensively the causes of febrile illness in northern Tanzania among patients sufficiently ill to require hospitalization . Febrile patients admitted to two hospitals were evaluated for a wide range of infectious etiologies using conventional standard diagnostic techniques .
This study was approved by the Kilimanjaro Christian Medical Centre ( KCMC ) Research Ethics Committee , the Tanzania National Institutes for Medical Research National Research Ethics Coordinating Committee , and Institutional Review Boards of Duke University Medical Center and the CDC . All minors had written informed consent given from a parent or guardian and all adult participants provided their own written informed consent . Moshi ( population , >144 000 ) is the administrative center of the Kilimanjaro Region ( population , >1 . 4 million ) in northern Tanzania and is situated at an elevation of 890 m above mean sea level . The climate is characterized by a long rainy period ( March–May ) and a short rainy period ( October–December ) [16] . Malaria transmission intensity is low [17] . KCMC is a consultant referral hospital with 458 inpatient beds serving several regions in northern Tanzania , and Mawenzi Regional Hospital ( MRH ) , with 300 beds , is the Kilimanjaro Regional hospital . Together KCMC and MRH serve as the main providers of hospital care in the Moshi area . In 2008 , KCMC admitted 22 , 099 patients and MRH admitted 21 , 763 patients . A study team that was independent of the hospital clinical team identified participants among infants and children admitted to KCMC from 17 September 2007 through 25 August 2008 , and among adolescents and adult admitted to KCMC and MRH in Moshi , Tanzania , from 17 September 2007 through 31 August 2008 . The methods of these studies have been described in detail elsewhere [7] , [8] . In brief , all admitted patients were screened for eligibility by study team members as soon as possible after admission and no later than 24 hours after admission . Infants and children aged from ≥2 months to <13 years , with a history of fever in the past 48 h or an axillary temperature ≥37 . 5°C or a rectal temperature of ≥38 . 0°C , and adolescents and adults aged ≥13 years and with oral temperatures of ≥38 . 0°C were invited to participate in the study . Patients admitted with known malignancy , renal failure , hepatic failure , bone marrow aplasia , trauma or surgery were excluded . A standardized clinical history and physical examination were performed on consenting patients by a trained clinical officer who was a member of the study team and who worked in parallel with the hospital admitting team . Provisional diagnoses by the hospital clinical team made independently of the study team were recorded and coded using the International Statistical Classification of Diseases and Related Health Problems , 10th Revision ( ICD-10 ) codes . Following cleansing of the skin with isopropyl alcohol and povidone iodine , blood was drawn from adults and adolescents for aerobic blood culture ( 5 mL ) and for mycobacterial blood culture ( 5 mL ) and from pediatric patients for a single aerobic blood culture ( 4 ml ) . In addition , blood was drawn for complete blood count , examination for blood parasites , and HIV antibody testing . Acute serum , plasma , and whole blood were archived on all participants . For patients found to be HIV seropositive , CD4-positive T lymphocyte count ( CD4 cell count ) and serum cryptococcal antigen level were also measured . HIV-seronegative patients were screened for the presence of acute HIV infection by polymerase chain reaction ( PCR ) for HIV-1 RNA . Urine was collected as soon as possible after admission for detection of urine antimicrobial activity and for antigen detection . A discharge form was completed at the time of discharge from the hospital that captured whether the patient died in hospital , the in-hospital management , and the discharge diagnoses coded using ICD-10 codes . The results of study investigations done in Moshi were provided immediately to the hospital clinical team to inform patient management . The results of investigations done at reference laboratories were provided to the hospital clinical team as they became available . The hospital clinical team was responsible for all aspects of patient management , following clinical judgment and use of locally adapted and developed treatment guidelines . All participants were asked to return to a study clinic 4–6 weeks after enrollment to provide a convalescent serum sample . To promote high levels of follow up , the study team provided a follow up appointment card prior to hospital discharge , made reminder telephone calls to participants during the week prior to the appointment , reimbursed travel expenses of returning participants , and when necessary a field worker made home visits . Laboratory evaluations were selected to reflect a range of infectious diseases that might occur in northern Tanzania . Priority was given to laboratory evaluations for infectious diseases that might require specific management . Data were entered using the Cardiff Teleform system ( Cardiff Inc . , Vista , Ca . , USA ) into an Access database ( Microsoft Corp , Va . , USA ) . When a diagnostic test was not applied to the whole cohort due lack of availability of an acute or convalescent sample , the proportion of confirmed cases in the tested group was extrapolated to the untested group by assuming that prevalence was the same in the tested group as in the untested group . Statistical analyses were performed with SAS version 9 . 1 software ( SAS Inc , Cary , NC ) .
Figure 1 summarizes participant screening , enrollment , and diagnostic testing . Of 870 febrile admissions to two hospitals in northern Tanzania enrolled in the study 484 ( 55 . 6% ) were female . Of participants , 467 ( 53 . 7% ) were infants and children with a median ( range ) age of 2 years ( 2 months - 13 years ) ; the remainder adolescents and adults with a median ( range ) age of 38 ( 14–96 ) years . Fifty seven ( 12 . 2% ) infants and children were HIV-infected compared with 157 ( 39 . 0% ) adolescents and adults . Among infants and children 34 ( 7 . 3% ) of 464 with hospital outcome data died; 2 ( 5 . 9% ) of those who died had invasive infections . Among adolescents and adults , 41 ( 10 . 3% ) of 398 with hospital outcome data died; 11 ( 26 . 8% ) of those who died had invasive infections . In hospital deaths could not be attributed to etiologies requiring serologic diagnosis due to the requirement for testing a convalescent serum sample . Table 1 shows the number of patients with acute and convalescent samples available for testing for each etiologic agent or group of etiologic agents . Not all tests could be applied to all participants because of limited volumes of sample for some participants , and by the lack of availability of convalescent serum for participants who died before the follow up visit or who did not return . The number of confirmed cases in each group is also shown . The proportion of febrile admissions attributed to each etiology is calculated . A complete sample set was available for 243–467 ( 52 . 0–100 . 0% ) infants and children and for 207–403 ( 51 . 4–100 . 0% ) adolescents and adults . Of 467 infants and children enrolled , malaria was the clinical diagnosis for 282 ( 60 . 4% ) , but was the actual cause of fever in 6 ( 1 . 3% ) . Bacterial and fungal bloodstream infections described in detail elsewhere [8] accounted for 16 ( 3 . 4% ) and 4 ( 0 . 9% ) febrile admissions , respectively , and were underrepresented on admission differential diagnoses . Bacterial zoonoses were identified among 49 ( 20 . 2% ) of febrile admissions; 5 ( 2 . 0% ) had brucellosis , 19 ( 7 . 7% ) leptospirosis , 7 ( 2 . 6% ) had Q fever , 18 ( 7 . 4% ) had spotted fever group rickettsioses , and none had typhus group rickettsioses . In addition , 34 ( 10 . 2% ) of participants had a confirmed acute arbovirus infection , all due to chikungunya ( Table 1 ) . No patient had a bacterial zoonoses or an arbovirus infection included in the admission differential diagnosis . Of 403 adolescents and adults enrolled , malaria was the clinical diagnosis for 254 ( 63 . 0% ) , but was the actual cause of fever in 8 ( 2 . 0% ) . Bacterial , mycobacterial , and fungal bloodstream infections described in detail elsewhere [7] accounted for 69 ( 17 . 1% ) , 14 ( 3 . 5% ) , and 21 ( 5 . 2% ) febrile admissions , respectively , and were underrepresented on admission differential diagnoses . Bacterial zoonoses were identified among 69 ( 33 . 3% ) of febrile admissions; 11 ( 5 . 3% ) had brucellosis , 21 ( 10 . 1% ) leptospirosis , 17 ( 7 . 9% ) had Q fever , 18 ( 8 . 7% ) had spotted fever group rickettsioses , and 2 ( 1 . 0% ) had typhus group rickettsioses . In addition , 21 ( 5 . 7% ) of participants had a confirmed acute arbovirus infection , all due to chikungunya ( Table 1 ) . No patient had a bacterial zoonosis or an arbovirus infection included in the admission differential diagnosis . Among all 870 participants , malaria was the clinical diagnosis for 528 ( 60 . 7% ) , but was the actual cause of fever in 14 ( 1 . 6% ) . By contrast , bacterial , mycobacterial , and fungal bloodstream infections accounted for 85 ( 9 . 8% ) , 14 ( 1 . 6% ) , and 25 ( 2 . 9% ) febrile admissions , respectively , and were underrepresented on admission differential diagnoses . Bacterial zoonoses were identified among 118 ( 26 . 2% ) of febrile admissions; 16 ( 13 . 6% ) had brucellosis , 40 ( 33 . 9% ) leptospirosis , 24 ( 20 . 3% ) had Q fever , 36 ( 30 . 5% ) had spotted fever group rickettsioses , and 2 ( 1 . 8% ) had typhus group rickettsioses . In addition , 55 ( 7 . 9% ) of participants had a confirmed acute arbovirus infection , all due to chikungunya ( Table 1 ) . No patient had a bacterial zoonoses or an arbovirus infection included in the admission differential diagnosis . The proportional etiology of febrile illness among study participants after extrapolating to the untested group is summarized in Figure 2 .
We demonstrate among hospitalized febrile patients in northern Tanzania that malaria is uncommon and over-diagnosed , while invasive bacterial , mycobacterial , and fungal infections are underappreciated . At the same time , the bacterial zoonoses leptospirosis , Q fever , and spotted fever rickettsioses , and to a lesser extent brucellosis , and the arbovirus infection chikungunya are common yet unrecognized causes of fever . Our findings point to important mismatches between clinical diagnosis and management with actual diagnoses that have major implications for patient care , disease control and prevention , and for judicious use of antimalarial medications . While the problem of malaria over-diagnosis has been appreciated for some time [14] , [15] , studies that comprehensively describe the causes of severe non-malaria fever requiring hospital admission beyond bloodstream infections have been lacking . The over-diagnosis of malaria results in inappropriate use of antimalarial medications and may be associated with higher case fatality rates among patients treated for malaria who do not have the infection [14] , [15] , [34] . While the underlying causes of the over-diagnosis of malaria are complex [35] , the lack of epidemiologic information about the importance of alternative infections and guidance on their management is likely to play a role . Our findings confirm the potential benefits of making reliable malaria diagnostic tests available at healthcare facilities and using the results as the basis for prescription of antimalarial medications [36] . When adopted , such an approach to malaria treatment would support the judicious use of antimalarials and would define the population of patients with nonmalaria fever . We found that the bacterial zoonoses , leptospirosis , Q fever , and spotted fever group rickettsioses , and to a lesser extent brucellosis , are major causes of febrile illness among patients sufficiently unwell to require hospitalization . That a group of neglected bacterial zoonoses are of major clinical and public health importance in sub-Saharan Africa is a new and paradigm-changing finding . For clinical practice , with the exception of leptospirosis that may be effectively treated with commonly prescribed antibacterials , patients with brucellosis , Q fever , and the rickettsioses are likely to leave hospital without specific treatment . In northern Tanzania where many rely on livestock for their health and economic wellbeing , Leptospira , Brucella , and Coxiella spp . also indirectly affect human health through their impact on animal fertility , growth , and survival . The control and prevention of the neglected bacterial zoonoses is likely to involve interventions that require the collaboration of human health experts with the animal and environmental health disciplines , an approach that is underdeveloped in many parts of the world . Clinical guidelines for management of febrile patients in low resource areas focus on the identification and treatment of malaria and bacterial sepsis [11]–[13] . Our findings suggest that there is a need to identify and incorporate guidance on when to use a tetracycline for treatment of Q fever or rickettsial infection and when to consider treatment for brucellosis . We have previously demonstrated that features of the clinical history and physical examination do not perform well for identifying fever etiology [7] , [8] , [21] , [26] , [30] . Therefore , improvements to treatment algorithms for febrile patients are likely to require the development and incorporation of reliable diagnostic tests that provide timely diagnostic information to clinicians [37] . Unfortunately , many rapid diagnostic tests for infections related to fever management other than malaria and HIV suffer from poor performance characteristics [38] , [39] . Lack of coordination among groups working on the various etiologies of febrile illness in low-resource areas has meant that sentinel studies that could provide much more comprehensive information on a wide range of responsible organisms instead have focused on only one or a small handful of etiologies . For example , a clinical trial evaluating the impact of pneumococcal conjugate vaccine on rates of Streptococcus pneumoniae bacteremia in a community has the potential to identify and report all bloodstream infections . Similarly , a study designed to estimate the incidence of typhoid fever to inform vaccine policy could collect acute serum along with the blood culture and , with subsequent collection of convalescent serum , would have the ability to estimate the incidence of leptospirosis and a range of other etiologic agents using conventional serologic methods [40] . However , resources for research have tended to be targeted to specific pathogens and researchers have struggled to leverage additional resources to address a broader range of organisms . Sentinel site studies seeking to understand the infectious causes of febrile illness in low-resource settings have utilized blood culture to highlight the importance of invasive bacterial and fungal infections [4] , [41] . Expanding laboratory evaluations to include serologic and molecular approaches to diagnosing infections requiring specific antimicrobial management such as the bacterial zoonoses brucellosis , leptospirosis , Q fever , and the rickettsioses adds considerable value [40] . Detection of infections of public health importance such as those caused by the arboviruses dengue , Rift Valley fever , and yellow fever can inform national control programs . Since considerable etiologic overlap exists between the syndromes of fever , acute respiratory tract infection , and diarrhea [42] , [43] , addressing these simultaneously in integrated sentinel studies would inform enhancements in empiric treatment guidelines and improvements in the accuracy of syndrome-based disease burden estimates . Our study had a number of limitations . While we examined a wide range of etiologies of fever , a large proportion of patients were undiagnosed suggesting that we failed to identify potentially important infections . The undiagnosed group is being investigated further using pathogen discovery approaches . Some of the diagnostic tests used in our study are less than 100% sensitive and specific and we did not test for every known pathogen . As a consequence , we probably underestimated the prevalence of some infections while misclassifying others that were falsely positive . Because a number of our diagnostic tests relied on the demonstration of a four-fold rise in antibody titer between the acute and convalescent serum sample , not all enrolled patients returned for collection of convalescent serum to have diagnoses confirmed . It follows that calculation and comparison of case fatality rate was not possible since those who died before the convalescent visit could not be confirmed cases . Incomplete diagnostic information meant that we had to extrapolate prevalence from the tested population to the untested population , potentially introducing bias . Similarly , instances of apparent infection with multiple agents were not accounted for in presentation of pie graphs . Inclusion of a well control group would have allowed the calculation of attributable fractions for individual pathogens , something that should be considered for future febrile illness research , especially in areas where malaria is endemic . Since considerable geographic variation in fever etiology is known to occur , the generalizability of our findings is uncertain . What is needed to support an integrated approach to the syndrome of fever in resource-limited areas ? First , fever should be recognized alongside pneumonia and diarrhea as a major clinical syndrome of public health importance . Achieving this is likely to require leadership from international institutions of public health and reappraisal of the way that the febrile illnesses are approached in burden of disease estimates . This could include estimating total morbidity and mortality from the syndrome of fever as a first step before attributing the associated illnesses and deaths to specific etiologies , much as is done for the other major syndromes [44] , [45] . Second , efforts are needed to bring together the diverse groups and disciplines currently working on the febrile illnesses to quantify the morbidity and mortality attributable to each major etiologic agent . Such integration could be facilitated by support for research efforts that study the syndrome of fever comprehensively as well as its etiologies individually , an approach that has been modeled by studies addressing the syndromes of pediatric pneumonia and diarrhea in developing countries [46] , [47] . Third , improved diagnostic services are urgently needed to establish disease burden estimates and patient management for the febrile illnesses in resource-limited areas [10] . Conventional diagnostic tests for some infections , such as leptospirosis , are complex . For example , the collection of both acute and convalescent serum samples may be required , and testing services may be available at only a few national or supra-national reference laboratories . Assays relying on convalescent samples cannot be used to estimate case fatality rates [21] , [26] . Conversely , simple , rapid tests applied to acute samples may have poor performance characteristics [38] . Finally , clinical studies , including clinical trials , are needed to test and improve clinical management algorithms for febrile patients . The goal should be to target antimicrobial therapy to those who need it and to avoid inappropriate use among patients who will not benefit . In this way , patient outcomes can be improved , health resources can be conserved , and disease prevention and control efforts for febrile conditions can be rationally resourced .
|
The syndrome of fever is caused by a large number of infectious diseases . Malaria is thought to have been declining in the tropics since 2004 . Increasing use of malaria diagnostic tests reveal a growing proportion of patients with fever who do not have malaria . While malaria diagnostic tests may be available , healthcare workers have few tools to diagnose causes of fever other than malaria . In order to identify major causes of fever other than malaria in northern Tanzania , we studied 870 patients with fever who were sufficiently ill to require admission to hospital . Malaria was uncommon and over-diagnosed , whereas invasive infections , including bloodstream infections , were underappreciated . Infections associated with animals such as brucellosis , leptospirosis , Q fever , and spotted fever group rickettsioses as well as viral infections transmitted by mosquitoes were common yet overlooked . We recommend that research on the syndrome of fever in resource-limited areas should focus on a wide range of potential causes . Animal-associated infections should be prioritized in patient management and disease control .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"bacteremia",
"bacterial",
"diseases",
"infectious",
"diseases",
"bloodstream",
"infections",
"q-fever",
"salmonellosis",
"neglected",
"tropical",
"diseases",
"rickettsia",
"brucellosis",
"leptospirosis"
] |
2013
|
Etiology of Severe Non-malaria Febrile Illness in Northern Tanzania: A Prospective Cohort Study
|
RNA–mediated transmission of phenotypes is an important way to explain non-Mendelian heredity . We have previously shown that small non-coding RNAs can induce hereditary epigenetic variations in mice and act as the transgenerational signalling molecules . Two prominent examples for these paramutations include the epigenetic modulation of the Kit gene , resulting in altered fur coloration , and the modulation of the Sox9 gene , resulting in an overgrowth phenotype . We now report that expression of the Dnmt2 RNA methyltransferase is required for the establishment and hereditary maintenance of both paramutations . Our data show that the Kit paramutant phenotype was not transmitted to the progeny of Dnmt2−/− mice and that the Sox9 paramutation was also not established in Dnmt2−/− embryos . Similarly , RNA from Dnmt2-negative Kit heterozygotes did not induce the paramutant phenotype when microinjected into Dnmt2-deficient fertilized eggs and microinjection of the miR-124 microRNA failed to induce the characteristic giant phenotype . In agreement with an RNA–mediated mechanism of inheritance , no change was observed in the DNA methylation profiles of the Kit locus between the wild-type and paramutant mice . RNA bisulfite sequencing confirmed Dnmt2-dependent tRNA methylation in mouse sperm and also indicated Dnmt2-dependent cytosine methylation in Kit RNA in paramutant embryos . Together , these findings uncover a novel function of Dnmt2 in RNA–mediated epigenetic heredity .
Experimental results on model animals ranging from Caenorhabditis and Drosophila to the mouse have recently provided support for a mode of epigenetic heredity distinct from the canonical Mendelian rules [1]–[10] . These findings may help in understanding unexpected epidemiological results showing paternal transmission of pathological states over several generations [11]–[13] and provide at least partial solutions to the ‘missing heritability’ problem raised by genomic analyses [11] , [14] . Several of the current experimental systems point to RNA as the transgenerational signalling molecule , sperm RNA [15] in the case of paternal heredity . One important example of RNA-mediated inheritance is provided by the mouse paramutation , where transcriptional activation of a locus is mediated by small non-coding RNAs ( sncRNAs ) . These epigenetic variations were first detected by the hereditary maintenance of the white-tail phenotype of the Kit mutation in Kit+/+ offspring of heterozygotes carrying an inactivated allele ( KittmlAlf1/+ ) , which was associated with an accumulation of aberrant Kit transcripts in germ cells [8] . These RNAs were thought to play a role in the transgenerational transfer of the phenotype , a conclusion strengthened by microinjection assays in naive fertilized eggs . More specifically , oligoribonucleotides with sequences of the transcripts and Kit-specific microRNAs generated the hereditary modification . Similarly , microinjection in eggs of microRNA miR-1 resulted in overexpression of its target Cdk9 and that of miR-124 in increased expression of Sox9 during the preimplantation period . The miR-1/Cdk9 paramutants developed cardiac hypertrophy [4] and the miR-124/Sox9 variants a giant phenotype and twin pregnancies [10] . In all three cases , the epigenetic variations were stable over at least three generations of outcrosses and paternal transmission was explained by the transfer of sequence-related molecules in the spermatozoal RNA fraction . A search for genes involved in paramutation led us to consider a role of the Dnmt2 methyltransferase in RNA mediated epigenetic inheritance . In contrast to other members of the Dnmt family , the Dnmt2 protein catalyses cytosine methylation in RNA substrates , an activity which was at first enigmatic , homozygous null mutants of Drosophila , Arabidopsis and mouse being viable and fertile under laboratory conditions [16] . Methylation by Dnmt2 was reported to protect tRNAs from cleavage under stress conditions [17] and to be involved in upholding steady state levels of tRNAs [18] . We now report that a homozygous loss-of-function mutation of the Dnmt2 locus prevents the appearance of epigenetic variants of the Kit and Sox9 loci . Our results indicate that the methyltransferase is not required for expression of the variant phenotype during development . Our data further indicate a Dnmt2-dependent initiation step and suggest a role for Dnmt2 in the homeostasis of sncRNAs in the early embryo .
The white tail and feet of the KittmlAlf/+ heterozygotes ( Figure 1A ) are immediately recognizable , thus allowing for quantitative studies on relatively large numbers of mice . A non-Mendelian mode of transmission detected in their progenies had initially allowed us to identify a hereditary epigenetic modification of expression of the Kit+ allele ( paramutation ) , which is determined by cognate sncRNAs [8] . We then initiated a search for genes that would affect the establishment and/or maintenance of the paramutated state and considered the Dnmt2 RNA methyltransferase as a possible candidate . We generated 129/Sv mice carrying the heterozygous Kit locus and a Dnmt2 null mutation [16] . The results of crosses between KittmlAlf1/+ , Dnmt2−/− parents are summarized in Figure 1A , with a more detailed presentation in Table 1 and Figure S1 . The Dnmt2+/+ control crosses yielded the expected frequency of Kit paramutants ( Kit+/+ genotype with the white-spotted phenotype of the mutant ) . In contrast , in the progeny of two Dnmt2−/− parents , segregation of the phenotypes strictly corresponded to the Kit genotype . Crosses with Kit+/+ , Dnmt2+/+ mice of the wild type , full tail color Kit+/+ , Dnmt2−/− offspring failed to restore the modified state . A role of the genetic background was excluded because the results were reproduced in C57BL/6 and in B6D2F1 hybrids ( Table S1 ) . The regular segregation of the Kit+ phenotype in Dnmt2−/− crosses could have been explained by the selective mortality of variant embryos during development . However , further analysis argued against this possibility . As shown in Table 1 ( Exp #2 ) , all the embryos generated in 10 crosses between KittmlAlf1/+ Dnmt2−/− males and Kit+/+ Dnmt2−/− females were transplanted at the one-cell stage into Dnmt2+/+ foster mothers and 75 living births were obtained from 77 transplants . None of the Kit+/+ progenies showed the variant phenotype under these conditions , thus excluding embryonic lethality . Genetic analysis identified an initial period of establishment of the epigenetic variation . In crosses between KittmlAlf1/+ , Dnmt2+/− males and either Dnmt2−/− or Dnmt2+/− females , a fraction of the Kit+/+ Dnmt2−/− offspring showed the white-spotted phenotype ( Table 2 ) . This Dnmt2-negative Kit paramutant progeny was generated with a frequency identical to that in Dnmt2+/+ crosses . However , when these mice were subsequently crossed to wild-type partners , they did not further transmit the white tail phenotype . In other words , the epigenetic state was initially maintained in the Dnmt2−/− genotype during somatic development but was heritable only from a parent with an intact Dnmt2 allele . We conclude that Dnmt2 activity is critical during parental gametogenesis and/or in fertilized eggs . The resulting change in Kit expression in early stem cells can then be maintained in melanoblast stem cells in a Dnmt2-independent manner and results in the defect in their migration during early development responsible for the pigmentation of the adult tail . Dnmt2 is known to be expressed in oocytes and preimplantation embryos [19] and we detected both Dnmt2 RNA and protein in fractionated male germ cells in spermatocytes , round and elongated spermatids ( Figure S3 ) . We also analysed the methylation patterns in mouse sperm of the C38 target site in two established Dnmt2 substrates . The results showed high levels of C38 methylation for tRNA ( Asp ) and tRNA ( Gly ) in sperm from wild type mice ( Figure S4 ) . This methylation was substantially reduced in sperm from Dnmt2−/− mice ( Figure S4 ) , which provided confirmation for the enzymatic activity of Dnmt2 in the male germline . Further support for a role of Dnmt2 in the inheritance of epigenetic variation was obtained from RNA microinjection experiments . We had previously shown that microinjection into naive fertilized eggs of either RNA extracted from KittmlAlf1/+ tissues , or the cognate microRNAs , or oligoribonucleotides with transcript sequences induced the heritable phenotype modification . We then used these assays to compare the efficiency of RNA preparations from Dnmt2−/− and Dnmt2+/+ KittmlAlf1/+ heterozygotes . The results showed that RNA from the brains and testis of Dnmt2-deficient Kit heterozygotes did not induce the modified phenotype ( Table 3 ) . In subsequent experiments , an oligoribonucleotide with a sequence from the Kit mRNA ( nt 2123–2150 , Table S2 ) also induced the white-spotted phenotype when microinjected into wild-type one-cell embryos ( Table 4 ) . We also tested a form of the same Kit oligoribonucleotide in which all cytosines were methylated . This RNA , indicated as ‘Kit2123–2150met’ in Table 4 , was more efficient in inducing the modified phenotype in Dnmt2+/+ embryos but inefficient in the Dnmt2-deficient background , indicating a requirement for Dnmt2 expression in the embryo . We conclude that , while methylation of the inducer RNA is required for optimal efficiency , the methyltransferase is still needed in the most early embryonic period . To extend our analysis to a second example of a mouse paramutation , we tested whether the lack of Dnmt2 would affect the epigenetic modulation of Sox9 which can be induced by microinjection of the cognate microRNA miR-124 and of Sox9 transcript fragments in Dnmt2+/+ embryos [4] . The miR-124/Sox9 variants were characterized by augmented numbers of blastocyst stem cells and , as a result , overgrowth of the embryo and adult body and frequent twin pregnancies . Following microinjection of miR-124 into Dnmt2−/− fertilized eggs , E7 . 5 embryos were identical to controls ( Figure 1B ) and not oversized as the Dnmt2+/+ Sox9 paramutants . We concluded that the paramutation of Sox9 is also dependent on Dnmt2 expression . Modified patterns of DNA methylation have been reported in various instances of epigenetic variation [ref . 20 for review] including the maize paramutation [21] . We used methylated DNA immunoprecipitation ( meDIP ) to determine the DNA methylation status of the Kit locus in testis DNA from wild type , KittmlAlf1/+ and Kit+/+ paramutant mice . Assays were developed for three distinct regions covering the Kit promoter , exon 2 and exon 14 , respectively ( Figure 2A ) . The results indicated only background levels of methylation in the promoter , and substantial methylation in the two intragenic regions ( Figure 2B ) . This pattern was observed for all three genotypes ( Figure 2B ) , indicating that the Kit paramutation is not associated with altered DNA methylation profiles of the locus – although we cannot exclude a localized change in an unknown element at a distance , as described for the b1 paramutation of maize [21] . In parallel experiments , we also used this approach to determine the DNA hydroxymethylation status of the locus and found that hydroxymethylation levels were invariably low in all genotypes and regions tested ( Figure 2B ) . The meDIP findings were subsequently validated by DNA bisulfite sequencing of testis DNA . The results demonstrated that the Kit promoter was completely unmethylated and that the exon 14 region was completely methylated ( Figure 2C ) . This pattern was again observed for all three genotypes ( Figure 2C ) , which further suggests that paramutation is not associated with an altered DNA methylation profile of the Kit locus . We also used RNA bisulfite sequencing to analyze the possibility that Dnmt2 might methylate Kit transcripts . To this end , we induced the Kit paramutation by microinjection of an oligoribonucleotide ( Kit2123–2150 ) into fertilized eggs obtained from either two Dnmt2−/− or two wild-type parents . In parallel , we also prepared control embryos that were injected with buffer . RNA was prepared from E9 . 5 embryos and methylation analysis was performed on the 45 cytosines of a region amplified from Kit RNA ( nt 2100–2336 ) that covers the inducer oligoribonucleotide ( nt 2123–2150 ) and overlaps the exon 14∶15∶16 junctions . The results revealed two closely associated cytosines ( cytosines #4 and #8 , Figure 3 ) that remained unconverted in a higher fraction of reads , specifically in the microinjected Dnmt2+/+ embryos . Methylation of mRNA by Dnmt2 has not been reported so far and it is possible that our results have been influenced by deamination artifacts . However , the same two methylation sites were identified in three independent biological replicates and were not observed in the control embryos or in the oligoribonucleotide-treated Dnmt2−/− embryos ( Figure 3 ) , which suggests that they might represent genuine methylation marks . Contamination by DNA was excluded by the spliced structure of the sequence . Furthermore , we also tested the methylation pattern of the corresponding genomic sequence . The results showed methylated CpG sites that were clearly distinct from the sites detected in RNA and that were not dependent on the Dnmt2 genotype ( Figure S5 ) .
The physiological role of the RNA methyltransferase activity of Dnmt2 has been enigmatic for a long period of time . Dnmt2-mediated tRNA methylation has recently been linked to tRNA stability [17] , [18] . However , the widespread occurrence of 5-methylcytosine in RNA [22] may reflect a variety of functions , most of which still remain to be identified . The recurrent general considerations on a regulatory role of noncoding RNAs [reviewed in refs 23] , [24] led us to consider a possible physiological function of the methyltransferase in epigenetic regulation . The three instances of RNA-mediated hereditary variation that we reported as paramutations provided suitable experimental models . We now report that Dnmt2 is required for establishment and hereditary transmission of the epigenetic variation at the Kit and Sox9 loci . This was first revealed for the visible color phenotype of the Kit variants , a most classical approach in genetics . It was further confirmed and extended to Sox9 by microinjection experiments . Our data show that the parental RNAs and synthetic oligoribonucleotide inducers of the epigenetic variations were inefficient in Dnmt2-negative embryos . Evidence for RNA methylation in the inducer oligonucleotide sequence was observed in embryos undergoing the Kit paramutation . Furthermore , while the modified Kit phenotype was never observed in Dnmt2 −/− homozygotes born from two parents with the same genotype , it was , however , expressed by genetically identical homozygotes when at least one of their parents was a Dnmt2-positive heterozygote ( Table 2 ) . We concluded that the protein is required only during the parental gametogenesis or in the early embryo and not at later developmental stages – except for subsequent transgenerational transfer . At least two general explanatory models can be considered for the absolute requirement in Dnmt2 in the establishment of the epigenetic change . One model would be based on the knowledge that tRNAs are bona fide substrates of Dnmt2 , and that tRNA fragments are highly abundant in mouse sperm [25] . Our data show that at least two tRNAs are methylated in mouse sperm in a Dnmt2-dependent manner ( Figure S4 ) , which raises the possibility that methylation-dependent processing of tRNAs [17] could result in the generation of paramutation-inducing sncRNAs . However , we have so far been unable to detect any recognizable phenotypes after microinjection of various tRNAs and tRNA fragments ( data not shown ) . A second model would consider that the inducer small RNAs are maintained only in the Dnmt2+/+ genotype , possibly because they are methylated or complexed with methylated tRNAs . Such a model would also account for the increased efficiency of the methylated synthetic oligoribonucleotides ( Table 4 ) . Current preliminary results suggest that exogenous small RNAs introduced in the early embryo are stably maintained only in Dnmt2-positive embryos , leading us to the hypothesis of a protection against endonucleolytic cleavage by methylation in a manner analogous to tRNAs [17] . A control of the maintenance of parental small RNAs at the maternal-zygotic transition would be reminiscent of the mechanisms that , at the same developmental stages , eliminate part of the parental mRNAs [26] . In such a model , the new individual would actively constitute its own set of functional RNAs , both large and small , from the parental stocks .
The experiments here described were carried out in compliance with the relevant institutional and French animal welfare laws , guidelines and policies . They have been approved by the French ethics committee ( Comité Institutionnel d'Ethique Pour l'Animal de Laboratoire; number NCE/2012-54 ) . KittmlAlf1/+ heterozygotes were maintained in parallel in the original 129/Sv genetic background and in C57BL/6×DBA/2 F1 hybrids ( B6D2 ) . The Dnmt2−/− homozygote [16] was kindly provided by T . Bestor . Originally maintained on a mixed genetic background , the mutation was backcrossed onto 129/Sv , C57BL/6 and B6D2 genetic backgrounds , in each case for more than ten generations . Genotypes were determined by PCR analysis of Neo and LacZ expression and by Southern blot hybridization using a genomic probe . Total brain and testis RNA and oligoribonucleotides with Kit and miRNA sequences were adjusted to a concentration of 1 µg/ml and microinjected into B6D2 fertilized eggs according to established methods of transgenesis [27] . Quality of RNA preparations from the mouse organs was checked by spectroscopic analysis using the Bioanalyzer 2100 apparatus ( Agilent Technologies , Santa Clara , CA ) ( Figure S2 ) . Oligoribonucleotides were obtained from Sigma-Prolabo ( sequences provided in Table S2 ) . Northern blot analysis was performed by standard methods [28] . For analysis , RNA was extracted with Trizol Reagent ( Invitrogen ) . Protein extracts for Dnmt2 Western blot were prepared from snap-frozen enriched germ cell populations obtained by homogenization in RIPA Buffer . Testicular fractions were purified by elutriation as described [29] . 20 µg of protein was fractionated onto a 15% denaturing SDS-polyacrylamide gel and transferred to nitrocellulose . The following antibodies were used for immunodetection: rabbit anti-Dnmt2 antibody ( Santa-Cruz , Rabbit polyclonal IgH sc-20702 , lot: B1903 ) 1∶100 and rabbit anti-ß-actin antibody ( Santa-Cruz , sc-47778 , lot: D0907 ) 1: 250 with peroxidase-coupled goat anti-rabbit secondary antibody ( Santa Cruz Biotechnology ) 1∶10 , 000 . Methylated DNA immunoprecipitation was performed as described previously [30] . Sequences of PCR primers are shown in Table S3 . DNA bisulfite sequencing analysis was performed by using the EpiTect Bisulfite Kit ( Qiagen ) , in combination with 454 sequencing technology ( Roche ) . Sequences of 454 bisulfite sequencing primers are shown in Table S3 and S4 . Methylation maps were generated by BISMA [31] . Analysis of cytosine methylation in Kit RNA was performed as described [32] , with minor modifications . RNA isolated using TRIzol ( Invitrogen ) was digested with DNase ( Promega ) . An aliquot of 6 µg of RNA dissolved in 20 µl of RNase-free water was mixed with 42 . 5 µl of “Bisulfite Mix” and 17 . 5 µl of “DNA Protect” buffer . The RNA was denatured at 70°C for 5 min , followed by 1 h incubation at 60°C . This cycle was repeated 5 times . RNA was isolated from the bisulfite reaction mix using the RNeasy Purification Kit ( Qiagen ) and treated with 0 . 5 M Tris-HCl , pH 9 at 37°C for 1 h . Finally , RNA was precipitated and further processed for sequencing , as described previously [32] . This included random barcoding during the reverse transcription reaction to confirm that the sequenced DNA molecules represented different RNA molecules . Sequences of PCR primers are shown in Table S3 and S4 . Sperm RNA was prepared as described [10] and analyzed as described previously [18] . Data are expressed as means ± s . e . m . A p-value of less than 0 . 05 was considered statistically significant .
|
The possibility of a mode of inheritance distinct from the Mendelian model has been considered since the early days of genetics . Only recently , however , suitable experimental models were created . We now see the development of new experimental systems detecting non-Mendelian inheritance in a variety of organisms , from worms to mice . We have previously shown that RNA molecules act as transgenerational inducers of epigenetic variations in mice . We are currently using Mendelian genetics to dissect the factors involved in RNA–mediated transgenerational signalling . By showing an absolute requirement for Dnmt2 in this process , our study extends our knowledge of this still somewhat enigmatic protein . We confirmed that RNA rather than DNA methylation by the protein is involved in epigenetic heredity , and our genetic results indicate a requirement during an early step in the reproductive process , between parental gametogenesis and the preimplantation stage .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"animal",
"models",
"biochemistry",
"rna",
"model",
"organisms",
"nucleic",
"acids",
"heredity",
"genetics",
"epigenetics",
"biology",
"mouse"
] |
2013
|
RNA–Mediated Epigenetic Heredity Requires the Cytosine Methyltransferase Dnmt2
|
The bivalent killed whole-cell oral cholera vaccine ( BivWC ) is being increasingly used to prevent cholera . The presence of O-antigen-specific memory B cells ( MBC ) has been associated with protective immunity against cholera , yet MBC responses have not been evaluated after BivWC vaccination . To address this knowledge gap , we measured V . cholerae O1-antigen MBC responses following BivWC vaccination . Adults in St . Marc , Haiti , received 2 doses of the BivWC vaccine , Shanchol , two weeks apart . Participants were invited to return at days 7 , 21 , 44 , 90 , 180 and 360 after the initial vaccination . Serum antibody and MBC responses were assessed at each time-point before and following vaccination . We observed that vaccination with BivWC resulted in significant O-antigen specific MBC responses to both Ogawa and Inaba serotypes that were detected by day 21 and remained significantly elevated over baseline for up to 12 months following vaccination . The BivWC oral cholera vaccine induces durable MBC responses to the V . cholerae O1-antigen . This suggests that long-term protection observed following vaccination with BivWC could be mediated or maintained by MBC responses .
Vibrio cholerae is the causative agent of cholera and responsible for approximately 1 . 3 to 4 million cases of diarrhea and 21 , 000 to 143 , 000 deaths , annually[1] . Large cholera epidemics occur frequently and are even more devastating when V . cholerae is introduced into an immunologically naïve population . Oral cholera vaccines ( OCVs ) are an essential component of the World Health Organization ( WHO ) strategic roadmap that aims to reduce 90% of cholera deaths by 2030[2] . There are three currently WHO prequalified , commercially available killed whole-cell OCVs . WC-rBS ( currently manufactured as Dukoral by Valneva ) is a whole-cell vaccine that consists of heat and formalin inactivated V . cholerae O1 derived from both the Inaba and Ogawa serotypes and includes recombinant cholera toxin B subunit ( CTB ) . A second bivalent vaccine , BivWC ( currently manufactured as Shanchol by Shantha Biotechnics ) , contains V . cholerae serogroups O1 and O139 but lacks the additional CTB antigen . The third vaccine , Euvichol by EuBiologics , is considered to be a bioequivalent to Shanchol . In 2013 , WHO created a stockpile of the BivWC vaccine to respond to cholera outbreaks worldwide and it has been increasingly utilized to reduce the burden of epidemic cholera[3 , 4] . Despite the increasing use of BivWC , there are still important questions about its immunogenicity , especially in areas outside the historically cholera-endemic areas of South Asia . While natural infection with V . cholerae induces long-term memory B cell ( MBC ) responses that are correlated with protection against V . cholerae infection[5] , the WC-rBS vaccine does not appear to induce significant MBC responses , despite a similar initial plasma antibody response[6 , 7] . This finding might account for the relatively short-lived protection afforded by the WC-rBS vaccine[8] . In contrast , data from recent clinical trials and epidemiologic studies demonstrate that BivWC vaccine likely affords longer lasting protection[9] . However , there are no data currently available on whether the BivWC vaccine induces a MBC response . To address this knowledge gap , we evaluated the development of cholera-specific MBC and serologic responses over the period of one year following BivWC vaccination in Haitian adults . The primary objective was to determine whether the BivWC vaccine , administered according to the currently recommended two-dose regimen 14 days apart , induced a MBC response in adults living in a cholera endemic region .
The study was a conducted in the Saint Nicolas Hospital in St . Marc , Haiti , an urban center in the Artibonite Department . We invited healthy adults , ages 18–60 years presenting to the Saint Nicolas Hospital outpatient clinic between 2015–2016 , to participate in this study . All participants provided a written informed consent prior to enrollment in the study . The protocol was reviewed and approved by the institutional review board of Partners HealthCare in Boston , Massachusetts and the Haitian National Bioethics Committee in Port-au-Prince , Haiti . All enrolled study participants were adults , over the age of 18 years . We excluded individuals if they had previously received OCV , were pregnant ( by a urine test ) , or had an active gastrointestinal disorder within the 7 days prior to enrollment . We administered two doses of BivWC vaccine ( Shanchol ) 14 days apart to the study participants in accordance with manufacturer’s recommendations . We collected venous blood samples at day 0 ( prior to vaccination ) , day 7 ( 7 days after the first dose of the vaccine ) , day 21 ( 7 days after the second vaccine dose ) , day 44 , day 90 , day 180 and day 360 . We isolated peripheral blood mononuclear cells ( PBMCs ) by density gradient centrifugation and cryopreserved in liquid nitrogen . We also collected plasma from the processed blood after centrifugation and stored them frozen at ≥ -80°C . We measured vibriocidal antibody titers at each time-point using guinea pig complement and V . cholerae O1 Ogawa PIC158 and Inaba PIC018 as the target organisms as previously described[10] . Briefly , the serum was 2-fold serially diluted in a 96 well plate 11 times ( until a final dilution of 1:10 , 240 ) . Each sample dilution was run in duplicate on the same plate . Diluted bacteria and complement ( final concentration 1:10 ) were added to each well prior to incubation on a shaker at 37°C for 1 hour ( 50 rev/min ) . Subsequently , 150μl of BHI media was added per well to grow bacteria strains for 2-4hrs at 37°C until it reached an optical density ( O . D . ) between 0 . 2 to 0 . 28 . We used a monoclonal antibody ( designated OSP-A2 ) that binds to the O-specific polysaccharide ( OSP ) moiety of LPS as a standard to monitor intra-assay variability between plates[11] . The O . D . s of each sample were averaged between wells within the same plate and experiments were rejected and repeated , if we did not see 50% killing of V . cholerae with the concentration of 31 . 25 to 62 . 5ng/mL of the monoclonal antibody . All time-point samples from the same participant were run on the same plate to minimize assay variations . The vibriocidal titer was defined as reciprocal of the highest dilution resulting in >50% reduction of the optical density of control wells without plasma . We used a standard enzyme-linked immunosorbent assay ( ELISA ) to measure plasma IgA , IgG and IgM antibody responses to V . cholerae -OSP as described previously[12 , 13] . OSP was purified and conjugated to bovine serum albumin ( BSA ) as previously described[14 , 15] . We coated ELISA plates ( MaxiSorp high affinity protein binding plates ) with either V . cholerae O1 Ogawa or Inaba OSP conjugated to BSA ( 1 μg/mL ) dissolved in 50mM carbonate buffer ( pH 9 . 6 ) . We then incubated plates with 100 μL of plasma ( diluted in 1:50 ) and bound antibody was detected with horseradish peroxidase-conjugated goat anti-human IgA , IgG or IgM ( dilution at 1:1000 ) and the color substrate , ABTS/H2O2 ( Sigma-Aldrich , St . Louis , MO ) . We measured absorbance values at 405 nm wavelength using kinetic readings ( milliabsorbance/second ) and SoftMax Pro software ( version 5 . 3 , Sunnyvale CA ) . We measured MBC responses to V . cholerae-specific antigens as previously described[16] . Briefly , we thawed and rested PBMCs overnight at 37°C , 5% CO2 , and then resuspended to a concentration of 5 x105 PBMC/ well in 24-well cell culture plates ( BD Biosciences , San Jose , CA ) . We then added a mixture of mitogens , optimized to stimulate antigen-independent proliferation and differentiation of memory B cells into antibody-secreting cells ( ASCs ) to some wells , while media without mitogens was added to ‘no stimulation’ control wells . The mitogens were added to a final concentration of 3 μg of CpG ( Toll-like receptor 9 agonist ) oligonucleotide/mL ( Operon , Huntsville , AL ) , 20 ng of B-cell activating factor ( BAFF ) / mL ( PeproTech , Rocky Hill , NJ ) and 5 ng of cytokine interleukin-15 per mL ( PeproTech , Rocky Hill , NJ ) [17] . We then incubated the cells at 37°C with 5% CO2 for 6 days . For the enzyme-linked immunosorbent spot ( ELISPOT ) assay , we coated nitrocellulose-bottom plates ( Mashan-4550; Millipore , Bedford , MA ) with 100μl of V . cholerae O1 Ogawa or Inaba , OSP conjugated to BSA , at a concentration of 10 μg/ml in 1x Phosphate-buffered saline ( PBS ) . We also coated plates with 100μl of affinity-purified goat anti-human IgG F ( ab ) 2 ( Jackson Immunology Research , West Grove , PA ) at a concentration of 5 μg/mL in PBS ( pH 7 . 4 ) , CtxB at 2 . 5 μg/mL ( Sigma Aldrich ) and keyhole limpet hemocyanin ( KLH , Pierce Biotechnology , Rockford , IL , 2 . 5 μg/mL ) . We used KLH as a negative control to detect non-antigen specific responses , and CtxB to assess for MBC responses that might result from exposure to natural V . cholerae infection . We used 20% of harvested PBMCs from each well to measure total IgG and IgA MBC and 80% to measure antigen-specific IgG and IgA MBCs . To detect MBCs , we used a dual color assay with horseradish peroxidase-conjugated mouse anti-human IgA ( Hybridoma Reagent Laboratory , Baltimore , MA ) and alkaline phosphatase-conjugated mouse anti-human IgG ( Southern Biotech , Birmingham , AL ) . We then developed the plates with 5-bromo-4-choloro-3-indolylphosphate-nitroblue tetrazolium ( BCIP/NBT , Sigma-Aldrich ) and 3-amino-9-ethylcarbazole ( AEC , Sigma Aldrich ) . Two individuals using a stereomicroscope independently quantified MBCs and the average of these determined the final number of antigen-specific and total IgG and IgA expressing MBCs per sample . As in previous studies , we excluded from our analysis samples that did not demonstrate sufficient stimulation by mitogens or a high number of negative control spots[6 , 16] . Specifically , samples which did not demonstrate a ≥ 4-fold increase in total IgG or IgA antibody secreting cells after stimulation , or those in which the KLH wells contained > 3 spots were excluded from the analysis . A responder was defined as an individual with detectable MBC after vaccination with no prior MBCs on day 0 or an individual with detectable MBCs on day 0 and a 50% increase of MBCs after vaccination . The percentage of responders is the combination of both these groups over the total number of vaccinees . Because measuring IgM MBC responses by ELISPOT results in a high number of non-specific spots in both V . cholerae and KLH wells and because of the limited number of available PBMCs , we instead measured V . cholerae O1 antigen-specific IgM antibodies in MBC supernatants by ELISA as described previously[18] . We collected MBC cell culture supernatants after 6 days of incubation from both stimulated and un-stimulated cultures and added a cocktail of protease inhibitors before freezing at -80°C . We measured the OSP-specific IgM responses in the supernatants using the ELISA method described above for plasma . In addition , we measured the total IgM in each supernatant to normalize the OSP-specific IgM antibody level to the total IgM concentration per well . We performed statistical analyses using STATA Version 14 ( StataCorp , LP , College Station , TX ) and Graph-Pad Prism ( Graph Pad Software , Inc . , La Jolla , CA ) . We expressed vibriocidal titers as geometric mean titers ( GMT ) with 95% confidence intervals . Antibody levels were expressed as a mean of ELISA units with standard error; and antigen-specific MBC responses were presented as a percentage of the total number of IgG or IgA MBCs . We used a paired t-test to compare responses between baseline ( pre-vaccination ) and subsequent time points .
A representation of the enrollment and follow-up of the study participants is shown in Fig 1 . The participants were enrolled in three separate cohorts , and the first set of 24 participants did not have a day 44 visit which is the reason for the lower number of participants evaluated at this time point . 93% of the participants returned for a minimum of 1 follow-up point , therefore a total of 68 individuals were included in subsequent statistical comparisons . Demographic features of the study participants are listed in Table 1 . No adverse events were reported related to vaccination of the study participants . Vibriocidal antibody response is a measurement of the total amount of antibodies targeting the V . cholerae bacteria and is commonly used to determine the level of prior exposure within an individual . Vibriocidal antibody responses following vaccination are shown in Fig 2 , with seroconversion rates listed in Table 2 . As expected , the baseline vibriocidal titers of the participants suggested a high level of previous exposure to V . cholerae; 18 of 73 ( 24 . 6% ) had a vibriocidal titer ≥ 80 for Ogawa , while 7 of 73 ( 9 . 6% ) had a vibriocidal titer ≥ 80 for Inaba . Consistent with our previous evaluations , over 80% of adults had a greater than fourfold increase in vibriocidal antibody titer against both serotypes following vaccination[19 , 20] . There is a significantly greater fold rise over baseline of Inaba vibriocidal antibody responses versus Ogawa on day 44 , as measured by a paired t-test ( p = 0 . 02 ) . The vibriocidal antibody response peaked on day 21 after 2 doses of vaccine and decreased in the subsequent time-points ( day 44 , 90 , 180 and 360 ) . However , in aggregate , the vibriocidal antibody titers remained significantly elevated over baseline titers even up to 1 year after vaccination for both the Inaba and Ogawa serotypes . Vibriocidal antibody titer geometric means and 95% confidence intervals for all time points are included in S1 Table . A small number of participants ( 2 for Ogawa and 1 for Inaba ) also had a fourfold or greater increase in vibriocidal titers by day 360 over day 180 suggesting possible re-exposure to V . cholerae during this period . O-specific polysaccharide ( OSP ) is the major antigen of immune responses that correlate to protection in V . cholerae infections . We determined antibody responses to the V . cholerae OSP antigen after each dose of BivWC vaccination and until day 360 ( Fig 3 ) . We observed a robust response increase in Ogawa-OSP IgG and remained significantly elevated over baseline up to a year after vaccination . In contrast , mean IgM responses peaked on day 21 , while IgA responses peaked on day 7; and there was no clear evidence of a sustained increase in circulating IgM or IgA OSP specific antibodies beyond day 44 . Similar to the vibriocidal responses , we observed an increase in responses on day 360 relative to day 180 raising the possibility of re-exposure to V . cholerae during this period . We also performed an additional analysis , excluding a subset of individuals with evidence of recent prior exposure to V . cholerae ( those individuals who had a day 0 vibriocidal titer >80 ) . Among this subset antibody responses remained significantly elevated over baseline within the same individual . However , vaccinees with serologic evidence of recent past exposure had a significantly increased rapid Ogawa-IgG antibody response compared to individuals without evidence of prior exposure after the first vaccine dose on day 7 ( p = 0 . 04 ) . We also consistently observed a lower antibody response targeting Inaba-OSP compared to Ogawa-OSP for the IgG isotype , by a paired t-test analysis of fold-rise antibody responses on days 21 , 44 , 90 and 180 ( p-values of 0 . 02 , 0 . 02 , 0 . 019 and 0 . 003 , respectively ) . Whereas fold-rise of IgA antibodies targeting Inaba-OSP is lower than Ogawa-OSP on day 7 ( p = 0 . 0004 ) . However , the differences between Ogawa and Inaba antibody increases were no longer significant when removing individuals with evidence of previous exposure . We measured the development of MBC responses in individuals to determine whether the BivWC vaccine is capable of stimulating these memory responses ( Fig 4 ) . V . cholerae Ogawa-OSP antigen-specific MBCs demonstrated the most robust increase over baseline . The majority of vaccinees had a V . cholerae Ogawa-OSP IgG MBC response , with 67% demonstrating a response at day 21 and 64% demonstrating a response at day 44 . In aggregate , these responses remained significantly elevated over baseline at 12 months post vaccination , the latest time point assessed . V . cholerae Inaba-OSP specific IgG-MBCs were also significantly elevated up to 6 months after vaccination . We also observed an increase in the number of circulating V . cholerae OSP specific IgA MBCs to both the Inaba and Ogawa serotypes , though these responses were detected in the circulation for a shorter duration of time after vaccination . Again , most vaccinees had detectable Ogawa IgA MBC responses following vaccination with a 51% responder frequency at day 21 and 60% at day 44 . In addition , V . cholerae Ogawa and Inaba-OSP IgM-MBC responses ( Fig 5 ) , as measured by ELISA in cell culture supernatants , ( see methods for more experimental rationale ) , were detected after vaccination . They were no longer elevated after day 90 for Ogawa and day 44 for Inaba . We performed multiple regression analyses between early vibriocidal titers and subsequent immune responses to determine whether vibriocidal titers would be predictive of later immunity . However , we did not find significant associations between vibriocidal titers at baseline , day 7 or day 21 to any of the subsequent immune responses we evaluated in this study . To evaluate whether the generation of MBC responses over the study period may have occurred from exposure to V . cholerae infection , we evaluated MBCs to the CTB protein for all individuals , at all-time points measured . CTB MBC responses were detected in 13 samples; 4% of total samples . When these samples were excluded from the overall analysis , it did not change the significance of MBC responses to vaccination at any time point over baseline . To address the impact of likely prior exposure on memory responses after vaccination , we also performed a subsequent analysis in which we removed the individuals with baseline vibriocidal titers of 80 and above ( S1 and S2 Figs ) . V . cholerae Ogawa-IgG MBC responses remain significantly elevated on days 21 , 44 and 90 , but the increase in MBCs were no longer significant for the later time points , day 180 or 360 . Similarly , V . cholerae Inaba-IgG MBC responses remain significantly elevated on days 21 and 44 but are no longer significantly elevated on days 90 and 180 . However , these later time points have the lowest sample numbers and the analysis could be underpowered .
We found that the BivWC vaccine administered using a standard two dose regimen in a cholera endemic area induced a robust MBC response to the V . cholerae O-specific polysaccharide ( OSP ) antigen . In addition , we found that levels of circulating Ogawa specific IgG MBC responses remained significantly increased through one year after vaccination . Our study demonstrated persistent increases in circulating vibriocidal antibody titers to both serotypes and OSP IgG antibody titers for at least one year following vaccination . This is an important finding since OSP specific MBC responses appear to be an immunological indicator of long-term protective immunity against cholera . For example , following natural infection with V . cholerae , MBC responses remain detectable longer than other immunologic responses after a V . cholerae infection[21]; and the presence of detectable circulating IgA or IgG OSP MBCs are predictive of protection against V . cholerae following infection[5] or vaccination[22] , even in exposed individuals without persistently elevated levels of circulating vibriocidal antibodies[5] . Our results differ from previous studies evaluating MBC responses to the WC-rBS vaccine [6 , 7 , 13] . These studies did not demonstrate a significant OSP MBC responses to the WC-rBS vaccination [6 , 7 , 13] . Interestingly , this difference in immunogenicity between OCVs appears to be consistent with data from both clinical trial and field effectiveness studies which demonstrate that the BivWC vaccine provides significantly longer lasting protection than the WC-rBS vaccine [9] . This difference in immunologic outcomes may reflect that CTB is a potent immunomodulatory agent . In mice , administration of CTB was found to suppress systemic immune responses , specifically dendritic cell priming Th2 responses and allergic inflammation [23 , 24] . CTB administered on mucosal surfaces also induces Treg cells , which would dampen immunological responses [25] . In this context , it is notable that the original trial of the precursor of the WC-rBS vaccine compared a whole cell formulation alone with a whole cell plus CTB formulation , and immunity following the CTB containing version waned more rapidly than it did for participants who were vaccinated with the whole cell only vaccine [26 , 27] . An alternate explanation for this difference is that it could be attributable to the antigen content of the WC-rBS vaccine versus the BivWC vaccine , as the LPS content of the BivWC vaccine is significantly greater[28] . We also observed longer lasting IgG antibody and memory B cell responses to the Ogawa-OSP antigen compared to Inaba-OSP after vaccination . Given that the Ogawa serotype has predominated in Haiti since 2010 , it is possible that higher levels of previous exposure may have primed the development of IgG MBC responses . In addition , we found shorter duration of IgG MBC responses when excluding individuals with high baseline vibriocidal titers , as a measurement of recent prior exposure . These findings suggest prior exposure may lead to increased immunogenicity and longer duration of memory responses after vaccination . Future studies should take into consideration previous exposure when evaluating or comparing the immunogenicity and effectiveness of OCVs in different settings . In conclusion , our study demonstrates the stimulation of long lasting O-specific polysaccharide antigen antibody and MBC responses by the BivWC vaccine in Haitian adults . The persistence of these immune responses was longer than expected given what has been observed in previous studies of the WC-rBS vaccine . Because MBC responses have been associated with protective immunity in previous studies [5 , 22] , these results might provide an immunologic marker and potential mechanistic basis for the longer term protection observed following BivWC vaccination in adults living in a cholera endemic area .
|
Oral cholera vaccines are being increasingly used throughout the world as a key component of cholera prevention programs . While several recent studies suggest oral cholera vaccines may provide durable protection , the potential mechanism that generates this long lasting immune memory and protection are unknown . Unlike antibody and antibody secreting cell responses , memory B cells are thought to be an important part of the immune responses because although these cells do not produce antibody , they are long lived and can be rapidly stimulated to produce antibodies upon re-exposure to infection . Previous studies have shown that memory B cell responses to the Vibrio cholerae O-antigen are associated with protection against cholera infection . In this study , we found that oral cholera vaccine generated long lasting antibody and memory B cell responses to the Vibrio cholerae O-antigen that remained elevated for 6 to 12 months . These findings show that oral cholera vaccination does induce a strong memory B cell response , which could play a role in the generation and maintenance of long-term protection following BivWC vaccination .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"cells",
"pathogens",
"vibrio",
"immunology",
"microbiology",
"vaccines",
"vibrio",
"cholerae",
"infectious",
"disease",
"control",
"antibodies",
"immunologic",
"techniques",
"memory",
"b",
"cells",
"bacteria",
"bacterial",
"pathogens",
"antibody",
"response",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"cholera",
"vaccines",
"medical",
"microbiology",
"proteins",
"microbial",
"pathogens",
"immunoassays",
"immune",
"response",
"biochemistry",
"antibody-producing",
"cells",
"cell",
"biology",
"b",
"cells",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2019
|
Bivalent oral cholera vaccination induces a memory B cell response to the V. cholerae O1-polysaccharide antigen in Haitian adults
|
Advances in optical neuroimaging techniques now allow neural activity to be recorded with cellular resolution in awake and behaving animals . Brain motion in these recordings pose a unique challenge . The location of individual neurons must be tracked in 3D over time to accurately extract single neuron activity traces . Recordings from small invertebrates like C . elegans are especially challenging because they undergo very large brain motion and deformation during animal movement . Here we present an automated computer vision pipeline to reliably track populations of neurons with single neuron resolution in the brain of a freely moving C . elegans undergoing large motion and deformation . 3D volumetric fluorescent images of the animal’s brain are straightened , aligned and registered , and the locations of neurons in the images are found via segmentation . Each neuron is then assigned an identity using a new time-independent machine-learning approach we call Neuron Registration Vector Encoding . In this approach , non-rigid point-set registration is used to match each segmented neuron in each volume with a set of reference volumes taken from throughout the recording . The way each neuron matches with the references defines a feature vector which is clustered to assign an identity to each neuron in each volume . Finally , thin-plate spline interpolation is used to correct errors in segmentation and check consistency of assigned identities . The Neuron Registration Vector Encoding approach proposed here is uniquely well suited for tracking neurons in brains undergoing large deformations . When applied to whole-brain calcium imaging recordings in freely moving C . elegans , this analysis pipeline located 156 neurons for the duration of an 8 minute recording and consistently found more neurons more quickly than manual or semi-automated approaches .
Optical neural imaging has ushered in a new frontier in neuroscience that seeks to understand how neural activity generates animal behavior by recording from large populations of neurons at cellular resolution in awake and behaving animals . Population recordings have now been used to elucidate mechanisms behind zebra finch song production [1] , spatial encoding in mice [2] , and limb movement in primates [3] . When applied to small transparent organisms , like Caenorhabditis elegans [4] , Drosophila [5] , and zebrafish [6] , nearly every neuron in the brain can be recorded , permitting the study of whole brain neural dynamics at cellular resolution . Methods for segmenting and tracking neurons have struggled to keep up as new imaging technologies now record from more neurons over longer times in environments with greater motion . Accounting for brain motion in particular has become a major challenge , especially in recordings of unrestrained animals . Brains in motion undergo translations and deformations in 3D that make robust tracking of individual neurons very difficult . The problem is compounded in invertebrates like C . elegans where the head of the animal is flexible and deforms greatly . If left unaccounted for , brain motion not only prevents tracking of neurons , but it can also introduce artifacts that mask the true neural signal . In this work we propose an automated approach to segment and track neurons in the presence of dramatic brain motion and deformation . Our approach is optimized for calcium imaging in unrestrained C . elegans . Neural activity can be imaged optically with the use of genetically encoded calcium sensitive fluorescent indicators , such as GCaMP6s used in this work [7] . Historically calcium imaging was often conducted in head-fixed or anesthetized animals to avoid challenges involved with imaging moving samples [4 , 8 , 9] . Recently , however , whole-brain imaging was demonstrated in freely behaving C . elegans [10 , 11] . C . elegans are a small transparent nematode , approximately 1mm in length , with a compact nervous system of only 302 neurons . About half of the neurons are located in the animal’s head , which we refer to as its brain . Analyzing fluorescent images of moving and deforming brains requires algorithms to detect neurons across time and extract fluorescent signals in 3D . Automated methods exist for segmenting and tracking fluorescently labeled cells during C . elegans embryogenesis [12] , and semi-automated methods are even able to track specific cells during embryo motion [13] , but to our knowledge these methods are not suitable for tracking neurons in adults . Generally , several strategies exist for tracking neurons in volumetric recordings . One approach is to find correspondences between neuron positions in consecutive time points , for example , by applying a distance minimization , and then stitching these correspondences together through time [14] . This type of time-dependent tracking requires that neuron displacements for each time step are less than the distance between neighboring neurons , and that the neurons remain identifiable at all times . If these requirements break down , even for only a few time points , errors can quickly accumulate . Other common methods , like independent component analysis ( ICA ) [15] are also exquisitely sensitive to motion and as a result they have not been successfully applied to recordings with large brain deformations . Large inter-volume motion arises when the recorded image volume acquisition rate is too low compared to animal motion . Unfortunately , large inter-volume brain motion is likely to be a prominent feature of whole-brain recordings of moving brains for the foreseeable future . In all modern imaging approaches there is a fundamental tradeoff between the following attributes: acquisition rate ( temporal resolution ) , spatial resolution , signal to noise , and the spatial extent of the recording . As recordings seek to capture larger brain regions at single cell resolution , they necessarily compromise on temporal resolution . For example , whole brain imaging in freely moving C . elegans has only been demonstrated at slow acquisition rates because of the requirements to scan the entire brain volume and expose each slice for sufficiently long time . At these rates , a significant amount of motion is present between image planes within a single brain volume . Similarly , large brain motions also remain between sequential volumes . Neurons can move the entire width of the worm’s head between sequential volumes when recording at 6 brain-volumes per second , as in [10] . In addition to motion , the brain also bends and deforms as it moves . Such changes to the brain’s conformation greatly alter the pattern of neuron positions making constellations of neurons difficult to compare across time . To track neurons in the presence of this motion , previous work that measured neural activity in freely moving C . elegans utilized semi-automated methods that required human proof reading or manual annotation to validate each and every neuron-time point [10 , 11] . This level of manual annotation becomes impractical as the length of recordings and the number of neurons increases . For example , 10 minutes of recorded neural activity from [10] , had over 360 , 000 neuron time points and required over 200 person-hours of proofreading and manual annotation . Here , we introduce a new time-independent algorithm that uses machine learning to automatically segment and track all neurons in the head of a freely moving animal without the need for manual annotation or proofreading . We call this technique Neuron Registration Vector Encoding , and we use it to extract neural signals in unrestrained C . elegans expressing the calcium indicator GCaMP6s and the fluorescent label RFP .
We introduce a method to track over 100 neurons in the brain of a freely moving C . elegans . The analysis pipeline is made of five modules and an overview is shown in Fig 1 . The first three modules , “Centerline Detection , ” “Straightening” and “Segmentation , ” collectively assemble the individually recorded planes into a sequence of 3D volumes and identify each neuron’s location in each volume . The next two modules , “Registration Vector Construction” and “Clustering , ” form the core of the method and represent a significant advance over previous approaches . Collectively , these two modules are called “Neuron Registration Vector Encoding . ” The “Registration Vector Construction” module leverages information from across the entire recording in a time-independent way to generate feature vectors that characterize every neuron at every time point in relation to a repertoire of brain confirmations . The “Clustering” module then clusters these feature vectors to assign a consistent identity to each neuron across the entire recording . A final module corrects for errors that can arise from segmentation or assignment . The implementation and results of this approach are described below . Worms expressing the calcium indicator GCaMP6s and a calcium-insensitive fluorescent protein RFP in the nuclei of all neurons were imaged during unrestrained behavior in a custom 3D tracking microscope , as described in [10] . Only signals close to the cell nuclei are measured . Two recordings are presented in this work: a new 8 minute recording of an animal of strain AML32 and a previously reported 4 minute recording of strain AML14 first described in [10] . The signal of interest in both recordings is the green fluorescence intensity from GCaMP6s in each neuron . Red fluorescence from the RFP protein serves as a reference for locating and tracking the neurons . The microscope provides four raw image streams that serve as inputs for our neural tracking pipeline , seen in Fig 2A . They are: ( 1 ) low-magnification dark-field images of the animal’s body posture ( 2 ) low-magnification fluorescent images of the animal’s brain ( 3 ) high-magnification green fluorescent images of single optical slices of the brain showing GCaMP6s activity and ( 4 ) high-magnification red fluorescent images of single optical slices of the brain showing the location of RFP . The animal’s brain is kept centered in the field of view by realtime feedback loops that adjust a motorized stage to compensate for the animal’s crawling . To acquire volumetric information , the high magnification imaging plane scans back and forth along the axial dimension , z , at 3 Hz as shown in Fig 2B , acquiring roughly 33 optical slices per volume , sequentially , for 6 brain-volumes per second . The animal’s continuous motion causes each volume to be arbitrarily sheared . Although the image streams operate at different volume acquisition rates and on different clocks , they are later synchronized by flashes of light that are simultaneously visible to all cameras . Each image in each stream is given a timestamp on a common timeline for analysis . Each of the four imaging streams are spatially aligned to each other in software using affine transformations found by imaging fluorescent beads . An example of the high magnification RFP recording is shown in S1 Movie . The animal’s posture contains information about the brain’s orientation and about any deformations arising from the animal’s side-to-side head swings . The first step of the pipeline is to extract the centerline that describes the animal’s posture . Centerline detection in C . elegans is an active field of research . Most algorithms use intensity thresholds to detect the worm’s body and then use binary image operations to extract a centerline [16–18] . Here we use an open active contour approach [19 , 20] to extract the centerline from dark field images with modifications to account for cases when the worm’s body crosses over itself as occurs during so-called “Omega Turns . ” In principle any method , automated or otherwise , that detects the centerlines should be sufficient . At rare times where the worm is coiled and the head position and orientation cannot be determined automatically , the head and the tail of the worm are manually identified . The animal’s centerline allows us to correct for gross changes in the worm’s position , orientation , and conformation ( Fig 3a ) . We use the centerlines determined by the low magnification behavior images to straighten the high magnification images of the worm’s brain . An affine transform must be applied to the centerline coordinates to transform them from the dark field coordinate system into the coordinate system of the high magnification images . Each image slice of the worm brain is straightened independently to account for motion within a single volume . The behavior images are taken at a lower acquisition rate than the high magnification brain images , so a linear interpolation is used to obtain a centerline for each slice of the brain volume . In each slice , we find the tangent and normal vectors at every point of the centerline ( Fig 3b ) . The points are interpolated with a single pixel spacing along the centerline to preserve the resolution of the image . The image intensities along each of the normal directions are interpolated and the slices are stacked to produce a straightened image in each slice ( Fig 3c ) . In the new coordinate system , the orientation of the animal is fixed and the gross deformations from the worm’s bending are suppressed . More subtle motion and deformation , however , remains . We further reduce shearing between slices using standard video stabilization techniques [21] . Specifically , bright-intensity peaks in the images are tracked between neighboring image slices . The coordinates of these peaks are used to calculate the affine transformations between neighboring slices of the volume using least squares . All slices are registered to the middle slice by applying these transformations sequentially throughout the volume . Each slice would undergo transformations for every slice in between it and the middle slice to correct shear throughout the volume . A final rigid translation is required to align each volume to the first volume of the recording . The translations are found by finding an offset that maximizes the cross-correlation between each volume and the initial volume . A video of straightening is shown in S1 Movie . Straightened images are used for the remaining steps of the analysis pipeline . Only the final measurement of fluorescence intensity is performed in the original unstraightened coordinated system . Before neuron identities can be matched across time , we must first segment the individual neurons within a volume to recovers each neuron’s size , location , and brightness ( Fig 3d and 3e ) . Many algorithms have been developed to segment neurons in a dense region [22 , 23] . We segment the neurons by finding volumes of curvature in fluorescence intensity in the straigthened volumes . After an initial smoothing , we compute the 3D Hessian matrix at each point in space and threshold for points where all of the three eigenvalues of the Hessian matrix are negative . This process selects for regions around intensity peaks in three dimensions . In order to further divide regions into objects that are more likely to represent neurons , we use a watershed separation on the distance transform of the thresholded image . The distance transform is found by replacing each thresholded pixel with the Euclidean distance between it and the closest zero pixel in the thresholded image . This approach is sufficient to segment most neurons . Occasionally neurons are missed or two neurons are incorrectly merged together . These occasional errors are corrected automatically later in the pipeline . Extracting neural signals requires the ability to match neurons found at different time points . Even after gross alignment and straightening , neurons in our images are still subject to local nonlinear deformations and there is significant movement of neurons between volumes . This remaining motion and deformation is clearly visible , for example , in S1 Movie . Rather than tracking neurons sequentially in time , the neurons in each volume are characterized based on how they match to neurons in a set of reference volumes . Our algorithm compares constellations of neurons in one volume to unannotated reference volumes and assigns correspondences or “matches” between the neurons in the sample and each reference volume . We modified a point-set registration algorithm developed by Jian and Vemuri [24] to do this ( Fig 4a ) . The registration algorithm represents two point-sets , a sample point-set denoted by X = {xi} and a reference point-set indicated by R = {ri} , as Gaussian mixtures and then attempts to register them by deforming space to minimize the distance between the two mixtures . In their implementation , each point is modeled by a 3D Gaussian with uniform covariance . Since we are matching images of neurons rather than just points , we can use the additional information from the size and brightness of each neuron . We add this information to the representation of each neuron by adjusting the amplitude and standard deviation of the Gaussians . The Gaussian mixture representation of an image is given by , f ( ξ , X ) =∑iAiexp ( −‖ξ−xi‖22 ( λσi ) 2 ) , ( 1 ) where Ai , xi , and σi are the amplitude , mean , and standard deviation of the i-th Gaussian . These parameters are derived from the brightness , centroid , and size of the segmented neuron , while ξ is the 3D spatial coordinate . A scale factor λ is added to the standard deviation to scale the size of each Gaussian . This will be used later during gradient descent . The sample constellation of neurons is then represented by the Gaussian mixture f ( ξ , X ) . Similarly , the reference constellation’s own neurons are represented as a f ( ξ , R ) . To match a sample constellation of neurons X with a reference constellation of neurons R , we use the non rigid transformation u : I R 3 ↦ I R 3 . The transformation maps X to u[X] such that the L2 distance between f ( ξ , u[X] ) and f ( ξ , R ) is minimized with some constraint on the amount of deformation . This can be written as an energy minimization problem , with the energy of the transformation , E ( u ) , written as E ( u ) = ∫ f ( ξ , u [ X ] ) - f ( ξ , R ) 2 d ξ + E Deformation ( u ) . ( 2 ) Note that the point-sets X and R are allowed to have different numbers of points . We model the deformations as a thin-plate spline ( TPS ) . The TPS transformation equations and resulting form of EDeformation ( u ) are shown in the methods . The minimization of E is found by gradient descent . Working with Gaussian mixtures as opposed to the original images allows us to model the deformations and analytically compute the gradients of Eq 2 making gradient descent more efficient . The gradient descent approach used here is similar to that outlined by Jian and Vemuri [25] . Since the energy landscape has many local minima , we initially chose a large scale factor , λ , to increase the size of each Gaussian and smooth over smaller features . Gradient descent is iterated multiple times with λ decreasing multiple times . After the transformation , sample points are matched to reference points by minimizing distances between assigned pairs using an algorithm from [14] . The matching is not greedy , and neurons in the sample that are far from any neurons in the reference are not matched . A neuron at xi is assigned a match vi to indicate which neuron in the set R it was matched to . For example if xi matched with rj when X is registered to R , then vi = j . If xi has no match in R , then vi = ∅ . The modified non-rigid point-set registration algorithm described above allows us to compare one constellation of neurons to another . In principle , neuron tracking could be achieved by registering the constellation of neurons at each time-volume to a single common reference . That approach is susceptible to failures in non-rigid point-set registration . Non-rigid point-set registration works well when the conformation of the animal in the sample and the reference are similar , but it is unreliable when there are large deformations between the sample and the reference , as happens with some regularity in our recordings . In addition , this approach is especially sensitive to any errors in segmentation , especially in the reference . An alternative approach would be to sequentially register neurons in each time volume to the next time-volume . This approach , however , accumulates even small errors and quickly becomes unreliable . Instead of either of those approaches , we use registration to compare the constellation of neurons at each time volume to a set of reference time-volumes that span a representative space of brain conformations ( Fig 4b ) , as described below . The constellation of neurons at a particular time in our recording is given by Xt , and the position of the i-th neuron at time t is denoted by xi , t . We select a set of K reference constellations , each from a different time volume Xt in our recording , so as to achieve a representative sampling of the many different possible brain conformations the animal can attain . These K reference volumes are denoted by {R1 , R2 , R3 , … , RK} . We use 300 volumes spaced evenly through time as our reference constellations . Each Xt is separately matched with each of the references , and each neuron in the sample , xi , t , gets a set of matches v i , t = { v i , t 1 , v i , t 2 , v i , t 3 , . . v i , t K } , one match for each of the K references . This set of matches is a feature vector which we call a Neuron Registration Vector . It describes the neuron’s location in relation to its neighbors when compared with the set of references . This vector can be used to identify neurons across different times . We find that 300 reference volumes creates feature vectors that are sufficiently robust to identify neurons in our recordings . What determines the optimal number of reference volumes ? As long as the reference volumes contain a representative sample of the space of brain conformation occupied during our recordings , the number of reference volumes needed to create a robust feature vector depends only on the size of this conformation space . Because the conformation space of a real brain in physiological conditions is finite , there exists some number of reference volumes beyond which adding more reference volumes provides no additional information . Crucially , the worm brain seems to explore this finite conformation space quickly relative to the time scales of our recordings . As a result , the number of required reference volumes should not depend on recording length , at least for the minutes-long timescales that we consider here . The neuron registration vector provides information about that neuron’s position relative to its neighbors , and how that relative position compares with many other reference volumes . A neuron with a particular identity will match similarly to the set of reference volumes and thus that neuron will have similar neuron registration vectors over time . Clustering similar registration vectors allows for the identification of that particular neuron across time ( Fig 4c and 4d ) . To illustrate the motivation for clustering , consider a neuron with identity s that is found at different times in two sample constellations X1 and X2 . When X1 and X2 have similar deformations , the neuron s from both constellations will be assigned the same set of matches when registered to the set of reference constellations , and as a result the corresponding neuron registration vectors v1 and v2 will be identical . This is true even if the registration algorithm itself fails to correctly match neuron s in the sample to its true neuron s in the reference . As the deformations separating X1 and X2 become larger , the distance between the feature vectors v1 and v2 also becomes larger . This is because the two samples will be matched to different neurons in some of the reference volumes as each sample is more likely to register poorly with references that are far from it in the space of deformations . Crucially , the reference volumes consist of instances of the animal in many different deformation states . So while errors in registering some samples will exist for certain references , they do not persist across all references , and thus do not effect the entire feature vector . For the biologically relevant deformations that we observe , the distance between v1 and v2 will be smaller if both are derived from neuron s than compared to the distance between v1 and v2 if they were derived from s and another neuron . We can therefore cluster the feature vectors to produce groups that consist of the same neuron found at many different time points . The goal of clustering is to assign each neuron at each volume to a cluster representing that neuron’s identity . Clustering is performed on the list of neuron registration vectors from all neurons at all times , {vi , t} . Each match in the vector , v i , t k , is represented as a binary vector of 0s with a 1 at the v i k - th position . The size of the vector is equal to the number of neurons in Rk . The feature vector {vi , t} is the concatenation of all of the binary vectors from all matches to the K reference constellations . For computational efficiency , a two-step process was used to perform the clustering: First agglomerative hierarchical clustering was used on the neurons from an initial subset of volumes to define the clusters . Next , neurons from all volumes at all times were assigned to the nearest cluster as defined by correlation distance to the clusters’ center of mass . Assignments were made in such a way so as to ensure that a given cluster is assigned to at most one neuron per volume . Details of this clustering approach are described in the methods . Each cluster is given a label {S1 , S2 , S3 , …} which uniquely identifies a single neuron over time , and each neuron at each time xi , t is given an identifier si , t corresponding to the cluster to which that neuron-time belongs . Neurons that are not classified into one of these clusters are removed because they are likely artifactual or represent a neuron that is segmented too poorly for inclusion . Neuron Registration Vector Encoding successfully identifies segmented neurons consistently across time . A transient segmentation error , however , would necessarily lead to missing or misidentified neurons . To identify and correct for missing and misidentified neurons , we check each neuron’s locations and fill in missing neurons using a consensus comparison and interpolation in a TPS deformed space . For each neuron identifier s and time t⋆ , we use all other point-sets , {Xt} to guess what that neuron’s location might be . This is done by finding the TPS transformation , ut→t⋆: Xt ↦ Xt⋆ , that maps the identified points from Xt to the corresponding points in Xt⋆ excluding the point s . Since the correspondences between neurons has already been determined , ut→t⋆ can be found by solving for the parameters from the TPS equation ( see methods ) . The position estimate is then given by ut→t⋆ [xi , t] with i selected such that si , t = s . This results in a set of points representing the set of predicted locations of the neuron at time t⋆ as inferred from the other volumes . When a neuron identifier is missing for a given time , the position of that neuron s is inferred by consensus . Namely , correct location is deemed to be the centroid of the set of inferred locations weighted by the underlying image intensity . This weighted centroid is also used if the current identified location of the neuron s has a distance greater than 3 standard deviations away from the centroid of the set of locations inferred from the other volumes , implying that an error may have occurred in that neuron’s classification . This is shown in Fig 5 , where neuron 111 is correctly identified in volume 735 , but the the label for neuron 111 is incorrectly located in volume 736 . In that case the weighted centroid from consensus voting was used . To assess the accuracy of the Neuron Registration Vector Encoding pipeline , we applied our automated tracking system to a 4 minute recording of whole brain activity in a moving C . elegans that had previously been hand annotated and published [10] . A custom Matlab GUI was used for manually identifying and tracking neurons . Nine researchers collectively annotated 70 neurons from each of the 1519 volumes in the 4 minute video . This is much less than the 181 neurons predicted to be found in the head [26] . The discrepancy is likely caused by a combination of imaging conditions and human nature . The short exposure time of our recordings makes it hard to resolve dim neurons , and the relatively long recordings tend to cause photobleaching which make the neurons even dimmer . Additionally , human researchers naturally tend to select only those neurons that are brightest and are most unambiguous for annotation , and tend to skip dim neurons or those neurons that are most densely clustered . We compared human annotations to our automated analysis in this same dataset . We performed the entire pipeline including detecting centerlines , worm straightening , segmentation , and neuron registration vector encoding and clustering , and correction . Automated tracking detected 119 neurons from the video compared to 70 from the human . In each volume , we paired the automatically tracked neurons with those found by manual detection by finding the closest matches in the unstraightened coordinate system . A neuron was perfectly tracked if it matched with the same manual neuron at all times . Tracking errors were flagged when a neuron matched with a manual neuron that was different than the one it matched with most often . The locations of the detected neurons are shown in Fig 6A . Only one neuron was incorrectly identified for more than 5% of the time volumes ( Fig 6B ) . The locations of neurons and the corresponding error rates are shown in Fig 6B . Neurons that were detected by the algorithm but not annotated manually are shown in gray . Upon further inspection , it was noted that some of the mismatches between our method and the manual annotation were due to human errors in the manual annotation , meaning the algorithm is able to correct humans on some occasions . GCaMP6s fluorescent intensity is ultimately the measurement of interest and this can be easily extracted from the tracks of the neuron locations across time . The pixels within an approximate 2 μm radius sphere around each point are used to calculate the average fluorescent intensity of a neuron in both the red RFP and green GCaMP6s channels at each time . This encompasses regions of the cell body , but excludes the neuron’s processes . The pixels within this sphere of interest are identified in the straightened RFP volume , but the intensity values are found by looking-up corresponding pixels in the unstraightened coordinate system in the original red- and green-channel images , respectively . We use the calcium-insensitive RFP signal to account for noise sources common to both the GCaMP6s and the RFP channel [10] . These include , for example , apparent changes in intensity due to focus , motion blur , changes in local fluorophore density arising from brain deformation and apparent changes in intensity due to inhomogeneities in substrate material . We measure neural activity as a fold change over baseline of the ratio of GCaMP6s to RFP intensity , Activity = Δ R R 0 = R - R 0 R 0 , R = I GCaMP 6 s I RFP . ( 3 ) The baseline for each neuron , R0 , is defined as the 20th percentile value of the ratio R for that neuron . Fig 7 shows calcium imaging traces extracted from new whole-brain recordings using the registration vector pipeline . 156 neurons were tracked for approximately 8 minutes as the worm moves . Many neurons show clear correlation with reversal behaviors in the worm .
The Neuron Registration Vector Encoding method presented here is able to process longer recordings and locate more neurons with less human input compared to previous examples of whole-brain imaging in freely moving C . elegans [10] . Fully automated image processing means that we are no longer limited by the human labor required for manual annotation . In new recordings presented here , we are able to observe 156 of the expected 181 neurons , much larger than the approximately 80 observed in previous work from our lab and others [10 , 11] . By automating tracking and segmentation , this relieves one of the major bottlenecks to analyzing longer recordings . The neuron registration vector encoding algorithm primarily relies on the local coherence of the motion of the neurons . It permits large deformations of the worm’s centerline so long as deformations around the centerline remain modest . Crucially , the algorithm’s time-independent approach allows it to tolerate large motion between consecutive time-volumes . These properties make it well suited for our neural recordings of C . elegans and we suspect that our approach would be applicable to tracking neurons in moving and deforming brains from other organisms as well . Certain classes of recordings , however , would not be well suited for Neuron Registration Vector Encoding and Clustering . The approach will fail when the local coherence of neuron motion breaks down . For example , if one neuron were to completely swap locations with another neuron relative to its surroundings , registration would not detect the switch and our method would fail . In this case a time-dependent tracking approach may perform better . In addition , proper clustering of the feature vectors requires the animal’s brain to explore a contiguous region of deformation space . For example , if a hypothetical brain were only ever to occupy two distinct conformations that are different enough that registration is not reliable between these two conformation states , the algorithm would fail to cluster feature vectors from the same neuron across the two states . To effectively identify the neurons in these two conformations , the animal’s brain must sample many conformations in between those two states . This way , discrepancies in registration arise gradually and the resulting feature vectors occupy a continuous region in the space of possible feature vectors . Note that a similar requirement would necessarily apply to any time-dependent tracking algorithm as well . We suspect that brain recordings from most species of interest meet these two requirements: namely neuron motion will have local coherence and the brain will explore a contiguous region of deformation space . Where these conditions are satisfied , we expect registration vector encoding to work well . Tracking in C . elegans is especially challenging because the entire brain undergoes large deformations as the animal bends . In most other organisms like zebrafish and Drosophila , brains are contained within a skull or exoskeleton and relative motion of the neurons is small . In those organisms , fluctuations in neuron positions take the form of rigid global transformations as the animal moves , or local non-linear deformations due to motion of blood vessels . We expect that this approach will be applicable there as well .
Transgenic worms were cultivated on nematode growth medium ( NGM ) plates with OP50 bacteria . Strain AML32 ( wtfIs5[Prab-3::NLS::GCaMP6s; Prab-3::NLS::tagRFP] ) was generated by UV irradiating animals of strain AML14 ( wtfEx4[Prab-3::NLS::GCaMP6s; Prab-3::NLS::tagRFP] ) [10] and outcrossing twice . Imaging is performed as described in Nguyen et al [10] . The worm is placed between an agarose slab and a large glass coverslip . The coverslip is held up by 0 . 006” plastic shims in order to reduce the amount of pressure on the worm from the glass , and mineral oil is spread over the worm to better match refractive indices in the space between the coverglass and the worm . The dark field image is used to extract the animal’s centerline while the fluorescent image is used for tracking the worm’s brain . Only the head of the worm is illuminated by the fluorescent excitation light and can be observed in the low magnification fluorescent image . The two low magnification videos and the RFP and GCaMP6 high magnification videos are aligned by imaging a slide of 4 μm “Tetraspeck” beads ( ThermoFisher ) that emit light in both red and green channels . We manually or automatically locate the beads from calibration images and use the bead positions to find affine transformations between each camera’s coordinate system . The affine parameters are found using a least squares fit on the coordinates of the beads in the image . Thin plate spline ( TPS ) transformations play an important role in error correcting and are also critical for the point set registration algorithm [24] . Given a set of n initial control points X = {xi} , and the set of transformed points , u[X] , the TPS transformation u can be written as u[X] = WU ( X ) + AX + t . The affine portion of the transformation is AX + t , while WU ( X ) is the non-linear part of the transformation from TPS . U ( X ) is an n × n vector with Ui , j=1‖xj−xi‖ and W is a 3 × n matrix . The elements of W , A and t are the parameters of the transformation u . These parameters are found in different ways dependant on context . During the error correction processing step , these parameters are fit by knowing both the the set of control points X and the location of the transformed points u[X] . In the context of the point set registration algorithm , u[X] incurs an energy penalty for deforming space given by EDeformation ( u ) = trace ( WUWT ) [24] . This cost is used in Eq 2 to determine the total energy of the transformation . Gradient descent is then used to determine the optimal TPS transformation parameters by minimizing the total energy of the transformation . Clustering is performed in two steps: hierarchical clustering and neuron classification . We chose to perform hierarchical clustering only on an initial subset of 800 volumes because hierarchical clustering can become prohibitively computationally intensive for larger datasets . The correlation distance , 1 − corr ( vm , vn ) , was used as the pairwise distance metric for clustering . Agglomerative hierarchical clustering was implemented using complete linkage with a distance cutoff of 0 . 9 . Clusters which are smaller than 40% of the number of subset volumes were removed . After the clusters were defined via hierarchical clustering , we then performed neuron classification . To classify neurons , we assigned neurons from every volume to the cluster with the nearest centroid . Only the best matched neuron in each volume is assigned to a cluster and only if the neuron is closer than some threshold distance , described below . If two or more neurons from a volume would otherwise be assigned to a single cluster , the closest neuron retains that classification and other neurons are unassigned . As a result , some putative neurons are not assigned to any cluster and at most one neuron per volume is assigned to any given cluster . The implementation of the algorithm is shown in Algorithm 1 . Algorithm 1 Clustering the Neuron Registration Vectors 1: input: Set of registration vectors V = {vi , t} 2: output: Cluster assignments for each of the vectors in V 3: procedure Cluster ( V ) 4: S = subset of V 5: subset_assignments = hierarchically cluster S with distance cutoff 0 . 9 6: cluster_list = unique ( subset_assignments ) 7: for each cluster in cluster_list do 8: If size ( cluster ) > 40% of volumes used then 9: cluster_center = average of S assigned to cluster 10: else 11: remove cluster 12: end if 13: end for 14: compute threshold from S 15: for each vi , t in V do 16: d = distances from vi , t to cluster_centers 17: if any ( d < threshold ) then 18: assign vi , t to closest cluster 19: end if 20: end for 21: for each volume in the recording do 22: for each cluster in cluster_list do 23: if multiple vi , t from volume assigned to cluster then 24: unassign all vi , t from the cluster except the closest one 25: end if 26: end for 27: end for 28: end procedure The threshold distance to determine whether a neuron is assigned to a cluster is calculated using a statistical analysis of the clusters generated by the initial hierarchical clustering so as to discriminate between neurons that are likely correctly or incorrectly assigned . The threshold is calculated as follows: For each neuron assigned during the initial clustering , we collect the distance between that neuron and the center of the cluster it was assigned to . The distribution of these distances is the “correctly assigned” distribution . In contrast , the null distribution is found by collecting the distances between each neuron and all clusters to which that neuron is not assigned . The threshold distance is set to be the largest distance for which a distance is more likely to be found in the “correctly assigned” distribution than the null distribution . The analysis was performed on Princeton University’s high-performance scientific computing cluster , “Della” primarily consisting of 240 nodes and 4288 cores , each with 2 . 4 GHz processors . Jobs were run on up to 200 cores simultaneously . Timing information for the steps listed in Fig 1 are described below and summarized in Table 1 .
|
Computer algorithms for identifying and tracking neurons in images of a brain have struggled to keep pace with rapid advances in neuroimaging . In small transparent organism like the nematode C . elegans , it is now possible to record neural activity from all of the neurons in the animal’s head with single-cell resolution as it crawls . A critical challenge is to identify and track each individual neuron as the brain moves and bends . Previous methods required large amounts of manual human annotation . In this work , we present a fully automated algorithm for neuron segmentation and tracking in freely behaving C . elegans . Our approach uses non-rigid point-set registration to construct feature vectors describing the location of each neuron relative to other neurons and other volumes in the recording . Then we cluster feature vectors in a time-independent fashion to track neurons through time . This new approach works very well when compared to a human .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"fluorescence",
"imaging",
"invertebrates",
"classical",
"mechanics",
"caenorhabditis",
"neuroscience",
"animals",
"animal",
"models",
"vector",
"construction",
"caenorhabditis",
"elegans",
"model",
"organisms",
"microscopy",
"experimental",
"organism",
"systems",
"damage",
"mechanics",
"dna",
"construction",
"molecular",
"biology",
"techniques",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"imaging",
"techniques",
"animal",
"cells",
"deformation",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"physics",
"gene",
"expression",
"and",
"vector",
"techniques",
"calcium",
"imaging",
"cellular",
"neuroscience",
"transmission",
"electron",
"microscopy",
"dark",
"field",
"imaging",
"cell",
"biology",
"neurons",
"electron",
"microscopy",
"nematoda",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"organisms"
] |
2017
|
Automatically tracking neurons in a moving and deforming brain
|
The Arabidopsis homeotic protein AGAMOUS ( AG ) , a MADS domain transcription factor , specifies reproductive organ identity during flower development . Using a binding assay and expression analysis , we identified a direct target of AG , GIANT KILLER ( GIK ) , which fine-tunes the expression of multiple genes downstream of AG . The GIK protein contains an AT-hook DNA binding motif that is widely found in chromosomal proteins and that binds to nuclear matrix attachment regions of DNA elements . Overexpression and loss of function of GIK cause wide-ranging defects in patterning and differentiation of reproductive organs . GIK directly regulates the expression of several key transcriptional regulators , including ETTIN/AUXIN RESPONSE FACTOR 3 ( ETT/ARF3 ) that patterns the gynoecium , by binding to the matrix attachment regions of target promoters . Overexpression of GIK causes a swift and dynamic change in repressive histone modification in the ETT promoter . We propose that GIK acts as a molecular node downstream of the homeotic protein AG , regulating patterning and differentiation of reproductive organs through chromatin organization .
During flower development , many key processes depend on tissue-specific regulation of gene expression achieved by the coordinated interplay of transcription factors . The classical ABC model was proposed nearly two decades ago to account for organ identity control in flower development [1] . The ABC model predicts that the combinatorial action of ABC floral homeotic genes controls floral organ identity . The ABC genes , A class for APETALA1 ( AP1 ) and APETALA2 ( AP2 ) , B class for APETALA3 ( AP3 ) and PISTILLATA ( PI ) , and C class for AGAMOUS ( AG ) , have been extensively studied and have been shown to encode transcription factors [2]–[8] . AG encodes a transcription factor of the MADS-domain protein family , and AG is necessary for the specification of stamens and carpels , the floral reproductive organs [5] , [6] . In ag-1 mutants , flowers undergo homeotic conversion to show a sepal-petal-petal reiteration instead of the normal sepal-petal-stamen-carpel structure . The complete lack of reproductive organs in ag-1 flowers places AG at the top of the hierarchy of genes controlling reproductive development . This conclusion is supported by microarray expression profiling of wild-type and ag mutant flowers showing that more than 1 , 000 genes are regulated downstream of AG [9] . Genome-wide studies by microarray using plant lines with controllable floral homeotic activities and chromatin immunoprecipitation ( ChIP ) led to the identification of direct target genes of the homeotic proteins [10]–[14] . AG directly regulates SPOROCYTELESS ( SPL , NOZZLE ) [15] , [16] to induce microsporogenesis , a process leading to pollen formation in Arabidopsis [12] . AG is expressed in developing stamens and regulates the expression of the catalytic enzyme DEFECTIVE IN ANTHER DEHISCENCE 1 ( DAD1 ) [17] to induce the biosynthesis of the phytohormone jasmonate , which is required for stamen maturation [18] . Along with SPL and DAD1 , genetic studies in Arabidopsis have revealed a large group of genes that are necessary for proper patterning and differentiation of reproductive organs . ETTIN ( ETT , AUXIN RESPONSE FACTOR3 ) acts redundantly with AUXIN RESPONSE FACTOR 4 ( ARF4 ) to participate in abaxial-adaxial axis patterning of the floral meristem and reproductive organs , as well as in the apical-basal patterning of the gynoecium [19]–[21] . LEUNIG ( LUG ) is implicated as a negative regulator of AG in petal primordia and also controls gynoecium fusion [22]–[24] . The YABBY family gene CRABS CLAW ( CRC ) is expressed preferentially in the abaxial side of carpels and is involved in specification of the gynoecium and nectaries [25] , [26] . JAGGED ( JAG ) and NUBBIN ( NUB ) , both encoding C2H2 zinc-finger transcription factors , function redundantly to promote proliferation of stamen and carpel primordia [27]–[29] . Another C2H2 zinc-finger transcription factor , KNUCKLES ( KNU ) , is involved in floral meristem determinacy and gametophyte specification , and its meristem expression is directly regulated by AG [30] , [31] . Nevertheless , the genetic pathways and networks leading to organogenesis are largely unknown , as are the molecular mechanisms that orchestrate the large number of transcriptional gene circuits downstream of AG . We report here the identification of GIANT KILLER ( GIK ) , a gene coding for an AT-hook type DNA binding protein , as a target of AG . GIK belongs to a protein family consisting of 29 members in Arabidopsis [32] , [33] . AT-hook DNA binding proteins may contribute to functional nuclear architecture by binding to the nuclear matrix [34]–[36] . The nuclear matrix is a putative structural component that remains inside the nucleus after removal of basic proteins and histones . AT-hook motifs bind to the minor grooves in duplex DNA of matrix attachment regions ( MARs ) of target DNA sequences [37] , [38] , a property that distinguishes them from common transcription factors that primarily bind to the major groove . MARs are stretches of characteristic AT-rich DNA sequences that tend to attain a single-stranded conformation through base unpairing ( thus , MARs are also called base unpairing regions , or BURs ) as a result of the torsional stress of the surrounding DNA [39] . MARs and AT-hook DNA binding proteins are believed to mediate anchoring of specific DNA sequences to the nuclear matrix , generating chromatin loop domains and possibly introducing structural changes in the chromatin [37] . In animals , the MAR binding protein SATB1 , which contains an AT-hook motif , has been implicated in tissue- or cell-type-specific regulation of multiple genes [40]–[44] . SATB1 may play a role in chromatin assembly and histone modification of nearby genes and may influence the transcription of multiple target genes . In plants , very little is known about developmental roles of AT-hook motif proteins , although close homologs of GIK have been isolated using yeast one-hybrid screening as promoter-binding proteins as well as from activation tagging screens [33] , [45]–[49] . We propose that GIK acts as a target of the floral homeotic protein AG and fine-tunes the expression of multiple genes involved in organ patterning and differentiation during reproductive development . Therefore , these data reveal one of the mechanisms by which homeotic genes regulate multiple downstream targets in plants .
We identified At2G35270 ( isolation name , 2-ATH; AHL21 [32] ) as a putative direct target of AG using bioinformatics screening ( Figure 1A , B ) of the Arabidopsis genome for potential AG binding sites and named it GIK ( as we found that it functions as a negative regulator of a gene whose name means “giant”; see below ) . First , we searched the entire Arabidopsis genome for the 16-bp consensus CArG box binding sequences of AG ( 5′-TTDCCWWWWNNGGHWW-3′ , D = A/T/G , W = A/T , N = A/T/G/C , H = A/T/C ) [50] , [51] and found 1 , 007 sites ( allowing one mismatch ) by utilizing the NCGR Patmatch program ( http://www . arabidopsis . org/cgi-bin/patmatch/nph-patmatch . pl ) . We then identified 110 genes located near the putative binding sites of AG ( within 3 kb upstream , 1 kb downstream , or in introns ) and tested their expression in wild-type and ag mutant flowers using RT-PCR ( Figure 1A , Figure S1 ) . Most of these genes were expressed in flowers . By comparing RNA in wild-type and ag-1 mutant flowers , we found that 24 of these genes ( 22% ) showed AG-dependent expression patterns ( Figure S1 , Table S1 ) ; of these , GIK , SHATTERPROOF2 ( SHP2 , AGL5 , At2g42830 ) [52] , [53] , and ATHB40 ( At4g36740 ) [54] , [55] showed rapid induction upon AG activation in an inducible AG activity line [12] ( unpublished data for ATHB40 , see below for GIK ) . We found a typical CArG box sequence located 732 bp downstream of the translational termination codon of GIK ( Figure 1B ) . After the series of experiments described below , we identified GIK as a direct target of AG . GIK encodes an AT-hook type DNA binding protein with an uncharacterized plant-and-prokaryote conserved domain ( Figure 1B ) . GIK transcripts were detected in roots , flowers , and leaves , with the highest expression in the roots , showing that GIK does not code for a flower-specific transcript ( Figure 1C ) . In ag mutant flowers , GIK expression was substantially reduced ( Figure 1C ) , suggesting that AG may be an upstream activator of GIK or that GIK is expressed in stamens and/or carpels , which are missing in ag-1 mutant flowers . To clarify these two possibilities , we used ag-1 plants that are transgenic for 35S::AG-GR as a post-translational AG activation system [12] and analyzed the expression of GIK following AG induction in developing flowers by RT-PCR and real-time PCR with GIK-specific primer sets ( Figure 1D–F; Table S2 for primer sequences ) . The transgenic line contains a gene coding for a fusion protein ( AG-GR ) of AG and the steroid binding domain of the rat glucocorticoid receptor ( GR ) on the ag-1 mutant background . Following application of the synthetic glucocorticoid dexamethasone ( DEX ) , AG-GR enters the nucleus and induces AG activity . GIK expression was upregulated 6 h after 10 µM DEX treatment compared to mock-treated inflorescences ( Figure 1D ) . The induction was observed even 2 h after DEX treatment ( Figure 1E ) . To exclude the possibility that GIK was induced indirectly by AG through an intermediate protein , we included the protein synthesis inhibitor cycloheximide [12] , [56] in our studies . DEX and 5 . 0 µM cycloheximide treatment induced GIK expression at a level comparable to DEX-only treatment , implicating a direct relationship between AG and GIK induction in developing flowers ( Figure 1E ) . In the time-course assay , GIK expression was upregulated by AG 4- and 16-fold at days 1 and 3 after AG induction , respectively ( Figure 1F ) . To test whether AG directly binds the Arabidopsis genome near GIK , we performed ChIP with a polyclonal antibody against AG ( anti-AG ) using 35S::AG-GR ag-1 inflorescences treated continuously with DEX . The primer pair hybridizing to the 3′ region of genomic GIK DNA containing a putative CArG box showed enrichment over a primer pair hybridizing to the coding region of GIK ( Figure 1G ) . The control experiment using untreated 35S::AG-GR ag-1 inflorescences did not show obvious enrichment ( Figure S2 ) . Our data suggest that AG directly activates GIK by binding to the region of the GIK CArG box in developing flowers . To determine whether AG is responsible for GIK expression in reproductive development , we examined GIK expression in inflorescences in detail using 3′ region of GIK cDNA as a probe for in situ hybridization . GIK transcripts were detected in inflorescence meristems , floral primordia , and developing flowers ( Figure 2A–G ) . GIK is expressed throughout floral primordia at stages 1 through 4 ( Figure 2A–D ) . At stage 6 and later , GIK expression is confined to reproductive organ primordia ( Figure 2D , E ) . At stages later than stage 10 , GIK is localized in developing ovules and anther locules ( Figure 2F , G ) . These data suggest that GIK expression is not fully dependent on AG ( as AG is expressed only in the central region of flower primordia after stage 2 ) but that expression later than stage 6 may depend on AG during reproductive development . In ag-1 inflorescences , although GIK is noticeably expressed in inflorescence meristems and floral primordia in early stages , GIK expression is considerably reduced in developing organs ( Figure 2H–J ) , which is consistent with the hypothesis that GIK is regulated by AG in reproductive organs . To examine GIK localization , we raised polyclonal antibodies against recombinant full-length GIK , the N-terminal and AT-hook domains . The antibody raised against full-length GIK detected a major protein band around 30 kDa in western blots ( Figure S3A ) , in agreement with the predicted GIK protein size of 29 . 1 kDa ( 285 residues ) . GIK was found in roots and flowers but not significantly in leaves ( Figure S3A ) . The level of GIK expression in roots and flowers corresponded well with our RT-PCR analysis of GIK transcripts in these tissues , where its expression in roots is much higher than in flowers ( Figure 1C , Genevestigator: www . genevestigator . ethz . ch , AtGenExpress: www . arabidopsis . org/info/expression/ATGenExpress . jsp ) , suggesting that GIK might play a role in root development . To determine whether GIK is a nuclear protein , we performed immunofluorescence staining using whole-mount seedlings and confocal microscopy . Staining was specifically detected in the nucleus ( Figure 2K–Q , Figure S3B ) and largely colocalized with two nuclear markers: the DNA dye TOPRO-3 and the trimethylguanosine cap of small nuclear RNA ( Figure 2K–Q ) . These results indicate that GIK is localized in the nucleoplasm . In addition , anti-GIK staining was distinguishable from both nuclear markers by a lack of GIK expression in heterochromatin chromocenters ( Figure 2K–M , arrowheads , regions observed as blue color in 2M ) and the nucleolus ( Figure 2O–Q , arrow in 2Q ) . To understand the role of GIK during flower development , we examined the effects of GIK overexpression . Over 20 transgenic plants from each transgenic line ( 35S::GIK and inducible 35S::GIK-GR-6HA ) were examined during flower development ( Figure 3A–G ) . At least three T1 35S::GIK plants showed reduced fertility with wide-ranging defects in reproductive development such as excessive outgrowth of stigmatic tissues ( Figure 3A , B ) , short valves ( Figure 3B ) , and excessive proliferation of a carpelloid organ at the lateral side of a pistil with exposed ovules ( Figure 3C ) . These phenotypes were largely recapitulated in nearly half of the T1 35S::GIK-GR-6HA lines after five DEX treatments ( Figure 3F , G ) . More than 90% of flowers from the induced 35S::GIK-GR-6HA plants showed severe reproductive defects such as excessive growth of stigmas or bipartite stigmas with outgrowth of ovules . In addition to defects in carpels , stamen development was occasionally affected , resulting in reduced male fertility ( unpublished data ) . Similar reproductive phenotypes were observed at low frequency ( 3% to 4% of their flowers ) in transgenic plants with a genomic copy of GIK , which showed 5- to 50-fold higher expression levels of GIK than wild-type plants ( unpublished data ) , indicating that the 35S::GIK and 35S::GIK-GR constructs provide high levels of GIK activity . These results show that overexpression of GIK strongly interferes with normal reproductive development . To examine whether GIK controls a subset of the known functions of AG , 35S::GIK was introduced into the ag-1 mutant plants . 35S::GIK did not rescue the ag-1 organ identity defects: no stamen- or carpel-like organs were observed in 35S::GIK ag-1 flowers , even though there was occasional sepal-sepal fusion ( Figure 3D ) . This observation suggests that the function of GIK is unlike many transcription factors that control cell differentiation or specification and that instead GIK may have a unique function in modulating gene expression downstream of AG . To further understand the role of GIK during flower development , a transposon insertion mutant of GIK ( http://genetrap . cshl . edu/TrHome . html , ET14389 ) was identified . It contains an insertion in the middle of the coding region , 450 bp from the start codon . Homozygous plants were verified by PCR-based genotyping and GIK expression analysis ( Figure S4A , B ) . Most of the flowers from gik homozygous mutants appeared normal without any gametophytic defects ( unpublished data ) , but a small number of flowers ( 22 of 800 ) showed various degrees of defects in stamen and carpel development ( Figure 3H , I , Figure S4C–E ) . Stamen development was impaired , which resulted in delayed dehiscence or indehiscence of anthers ( Figure 3H ) . In some cases , the filaments of the stamens were branched and had ectopic anther formation , and anthers were partially transformed into petal-like structures ( Figure 3I , Figure S4C , D ) . None of the defects were observed in wild-type plants grown under the same conditions . To examine whether the mutant phenotypes were caused by loss of GIK activity , we generated RNA interference ( RNAi ) silencing lines using the 3′ end of the GIK coding region ( Figure 3J , K ) . In 5 of 29 independent T1 RNAi plants , we observed similar defects of immature anthers and branching of stamen filaments at a similarly low frequency in T1 and T2 generations . We confirmed that the GIK transcripts were significantly reduced in flowers of GIK RNAi plants ( unpublished data ) . To examine whether GIK has redundant functions with other GIK-like genes [32] , [33] , we produced an RNAi silencing construct for the highly similar gene At4g17800 ( 67% amino acid identity ) and created the transgenic plants on the gik mutant background . However , we did not observe any obvious enhanced effects in the transgenic plants ( unpublished data ) . The GIK loss-of-function defects , albeit not at a high frequency , suggest some level of participation by GIK in reproductive development as a component of a fine-tuning mechanism . Because GIK overexpression phenotypes of outgrowth of stigmatic tissues , short valves , and bipartite stigmas with ectopic ovule formation ( Figure 3A–C , F , and G ) closely resemble loss-of-function phenotypes of the previously identified ettin ( ett , meaning “giant” ) mutants [19]–[21] , [57] , we examined whether there is a functional link between GIK and ETT . ETT encodes a member of an auxin response factor family of DNA binding proteins , and loss of ETT activity results in severe reproductive defects [20] , [57] . First , we crossed 35S::GIK-GR-6HA plants with the weak ett-3 mutant . Overexpression of GIK in the heterozygous and homozygous backgrounds of the weak ett-3 allele showed strong ett mutant phenotypes ( Figure S5 ) , suggesting an epistatic interaction of GIK overexpression with ETT . Next , we compared the expression patterns of GIK and ETT in wild-type reproductive organs in detail using in situ hybridization analysis ( Figure 4A–H ) . At floral stages 7–12 , GIK and ETT exhibited complementary expression patterns in the developing reproductive organs ( GIK , Figure 4A–D; ETT , Figure 4E–H ) . GIK is predominantly expressed in the adaxial part of the developing carpels and locules of stamens ( Figure 4A–C ) . On the other hand , ETT is expressed in the abaxial part of the developing carpels and in the vasculature of the stamens ( Figure 4E–G ) . In the developing ovules , complementary expression of GIK and ETT was also apparent ( Figure 4D , H ) : GIK was mainly expressed in funiculi , outer integuments , and the chalazal megaspores ( Figure 4D ) , whereas ETT expression was restricted to inner integuments and the nucellus ( Figure 4H ) . Next , we compared ETT expression in wild-type and 35S::GIK plants ( Figure 4I–L ) . ETT signals in the 35S::GIK flowers were lower than in the wild-type flower sections when stained on the same slide ( compare Figure 4I and J , 4K and L ) . Because the overexpression phenotype of GIK can be interpreted as the repression of GIK by ETT , we tested this possibility by examining GIK expression in ett mutant flowers . However , GIK expression was not upregulated in ett mutant flowers ( Figure S6 ) . Taken together , these results suggest that GIK can negatively regulate ETT , but not vice versa . Next , we examined ETT expression in the flowers of ag-1 mutants and inducible AG lines . ETT expression was only observed in the abaxial sides of early organ primordia and was not maintained in maturing organs in an ag mutant background ( Figure 4M ) . This indicates that late ETT expression , which can be modulated by GIK , requires AG activity . In ag-1 35S::AG-GR inflorescences , ETT expression was reduced to 80% of the initial level at 1 d after AG induction and then upregulated from day 2 onwards ( Figure S7 ) . These results suggest that ETT expression is positively regulated by AG , but at the same time , negatively modulated by GIK . To examine the regulatory effects of GIK on ETT in detail , time-course analysis of ETT expression was performed with 35S::GIK-GR-6HA transgenic plants with inducible GIK activity using real-time PCR . ETT expression was downregulated 4 h after a single DEX treatment that induces GIK activity , reached its lowest level at 8 h , and then returned to pretreatment levels ( Figure 4N ) , suggesting that induced GIK activity rapidly repressed ETT expression and that the ETT repression requires continuous GIK expression . Furthermore , we quantitatively measured ETT expression levels in gik mutant flowers . ETT was upregulated about 1 . 8 times in gik mutant flowers as compared to wild-type ( Figure 4O ) . These results suggest that GIK functions as an upstream negative modulator of ETT at certain floral stages . GIK contains an AT-hook DNA binding motif , which binds to the MAR of DNA sequences [36] , [38] . To examine how GIK controls ETT expression , we first examined whether GIK is a bona fide nuclear matrix-bound protein . Because the endogenous expression level of GIK in inflorescences is low ( Figure 1C , Figure S3A ) , we used inflorescences from 35S::GIK-GR-6HA plants . We isolated the nuclei from the inflorescences of DEX-treated 35S::GIK-GR-6HA plants ( the inflorescences were harvested 4 h after DEX treatment ) and then purified the matrix fraction by DNaseI treatment and extensive washing with high-salt buffer , which removes basic proteins and histones [58] , [59] . The total nuclear protein and the matrix fraction were probed with anti-HA that recognizes GIK-GR-6HA protein ( Figure 5A ) . A strong GIK signal was observed in the matrix fraction of the nuclei . In comparison , AG ( as a control ) was mostly washed away during the extraction processes , and only a faint signal was detected in the matrix fraction on the same membrane ( Figure 5A ) . This suggests that GIK is associated with the nuclear matrix . Next , to examine whether there are putative binding sites for GIK in the ETT promoter , we identified MARs in the upstream genomic region of ETT using SMARTest Software ( Figure 5B ) [60] . To test whether GIK can bind the putative MARs in the ETT promoter region , we expressed a truncated GIK with an intact AT-hook motif in E . coli and checked for its binding to an ETT putative MAR probe . We detected binding of the ETT probe to the GIK AT-hook domain ( Figure 5C , GIK-AT ) . The binding activity was reduced when one of the conserved binding regions , Arg-Gly-Arg-Pro ( Figure 1B ) [36] of the AT-hook domain , was mutated to Arg-Gly-Lys-Pro ( Figure 5C , GIK-MUT ) , suggesting that the wild-type AT-hook motif binds to the predicted MAR in the ETT promoter in vitro . To examine whether GIK binds to the putative MARs of the ETT promoter in vivo , we performed a ChIP assay using inflorescences from 35S::GIK-GR-6HA plants . The plants were treated with DEX , and the inflorescences were harvested 4 h later . Nuclear proteins were solubilized by sonication and immunoprecipitated with anti-HA . The putative MARs of the distal ETT promoter , especially the region represented by primer set P2 , showed clear enrichment ( Figure 5D ) . In contrast , neither the region that is close to one of the predicted MARs represented by primer set P4 nor the control showed enrichment ( Figure 5D ) . To examine whether endogenous GIK binds the putative MARs of the ETT promoter in a non-transgenic context , we repeated the ChIP experiment using wild-type inflorescences and the polyclonal anti-GIK . The result , albeit with some differences in the fold enrichment , indicated that GIK binds to the putative MARs of the distal ETT promoter in vivo ( Figure S8 ) . To evaluate whether the binding of GIK to the putative MARs of the ETT promoter is necessary for ETT regulation , we performed ETT promoter-reporter analysis ( Figure 5E , F , Figure S9 ) . We generated transgenic reporter lines in which the major MAR ( represented by primer sets P1 and P2 ) , located at distal part of the ETT upstream genomic region , was deleted ( pETTΔMAR::GUS ) ( Figure 5F ) . As a control , the upstream genomic region of ETT inclusive of all MARs ( pETT::GUS ) was fused with a GUS reporter gene and the inflorescences were stained ( Figure 5E ) . Expression of GUS in T1 pETTΔMAR::GUS transgenic lines was comparable or slightly weaker compared with that of pETT::GUS lines ( Table S3 ) . This result suggests that ETT expression is normal even after the deletion of these 5′ distal regions , and that the deleted regions may not contain regulatory elements or may contain both positive and negative regulatory elements for transcription . These reporter lines were crossed with the 35S::GIK-GR-6HA plants to test their responsiveness to ectopic GIK activation . There was a gradual reduction of GUS activity in response to continuous DEX treatment in plants transgenic for the construct with a full-length ETT promoter ( pETT::GUS ) in a time-dependent manner ( Figure 5E ) . At day 3 and later , GUS staining was barely detectable . In contrast , the pETTΔMAR::GUS reporter line was less responsive to GIK ( Figure 5F ) . To exclude the possibility that the no responsiveness is due to positional effects of an insertion site , we repeated the experiments using an independent line and confirmed that pETTΔMAR::GUS reporter line does not respond to GIK activity ( Figure S9A ) . To quantify this MAR-dependent repression of GUS activity by GIK , we carried out time-course GUS reporter gene expression analysis using quantitative real-time PCR ( Figure S9B , C ) . In agreement with the reduction in GUS staining , GUS expression in the pETT::GUS line was significantly downregulated at days 3 and 4 after GIK induction , respectively ( Figure S9B ) . In contrast , in the pETTΔMAR::GUS reporter line , there was no significant reduction of GUS expression at day 4 , and in fact a slight increase was seen at day3 ( ∼1 . 3-fold ) after GIK induction ( Figure S9C ) . These results suggest that repression of ETT by GIK requires the sequence containing the distal putative MARs of the ETT promoter . To examine whether repression of ETT is associated with any known epigenetic histone modifications , we performed a ChIP assay using antibodies against modified histones in wild-type and 35S::GIK backgrounds ( for details , see Materials and Methods ) . One of the repressive marks , dimethylated Lys 9 of histone H3 [61] , was found to be specifically enriched in the 35S::GIK background in the ETT upstream region ( Figure S10 ) . To gain further insight into the change in H3K9 dimethylation , we performed a time-course ChIP analysis using inflorescences of 35S::GIK-GR-6HA plants treated one time with 10 µM DEX . We observed a rapid increase in H3K9 dimethylation at the distal portion of the putative MAR within 2 h of GIK induction , especially in the region represented by primer set P2 ( Figure 5G ) . At 4 h post-induction , the increase in H3K9 dimethylation reached a maximum , with a 3- to 4-fold increase in the dimethylation level in the ETT upstream region ( Figure 5G ) . This change in dimethylated H3K9 was relatively rapid and dynamic: at the 8 h time point , the level was comparable to that at time 0 . ETT transcript levels were reduced to their lowest levels at the 8 h time point after GIK induction ( Figure 4N ) . This result suggests that the GIK-mediated ETT change requires continuous GIK activity and that the repression is closely associated with a dynamic change in the extent of H3K9 dimethylation in the ETT upstream region . To account for the pleiotropic phenotypes conferred by overexpression and loss of function of GIK ( Figure 3 ) , we examined a panel of reported Arabidopsis genes involved in reproductive development for their expression responses to GIK using real-time PCR ( Figure 6A–D and Table S4 ) . Many genes including LUG , which is a putative repressor of AG and whose loss of function leads to bipartite stigmas [22]–[24] , showed no clear changes in expression upon GIK activation in the time-course experiments using 35S::GIK-GR-6HA inflorescences ( Figure 6D , Table S4 ) . However , expression of CRC , JAG , and KNU decreased significantly after GIK induction ( Figure 6A–C ) . To determine whether GIK directly regulates CRC , JAG , or KNU , we first examined transcriptional repression by including the protein synthesis inhibitor cycloheximide ( Figure S11 ) . DEX with cycloheximide treatment repressed ETT , CRC , JAG , and KNU expression in 2 h at a level comparable to DEX-only treatment , indicating that transcriptional repression by GIK does not require de novo protein synthesis ( Figure S11 ) . In the upstream region of each of these genes , one to three predicted MARs were identified using SMARTest ( Figure 6E–G ) . The prediction made by the SMARTest program could contain false-positive and false-negative results ( Figure 5D ) [60] . To validate the SMARTest prediction , we performed ChIP experiments and showed apparent enrichment using primer sets that detect some of the putative MAR regions of these target genes ( Figure 6E–G ) . In the CRC promoter , there are two predicted MARs . Both the distal and proximal putative MARs showed a clear enhanced binding compared with control , whereas primer set P2 , which amplifies the 3′ region of the distal putative MAR , showed no clear enrichment ( Figure 6E ) . In the JAG promoter , the most distal of the three putative MARs showed the strongest enrichment ( Figure 6F ) . The 5′ transcribed region of JAG showed an unexpectedly high enrichment , which may indicate that an unpredicted MAR site is located in the transcribed region of JAG . In the KNU promoter , there was only one predicted MAR , and the enrichment index showed a bell-shaped distribution centered on the binding site ( Figure 6G ) . These results suggest that ectopically expressed GIK binds directly to the putative MARs of these target genes and represses their transcription . To determine whether endogenous GIK is involved in the regulation of CRC , JAG , or KNU , we examined the expression of these genes in the gik mutant background . Real-time PCR using flowers from gik homozygous plants showed that JAG and KNU were relatively highly expressed in the gik mutant ( Figure 6H ) . Expression of CRC was slightly increased , but expression of LUG was not changed in the gik mutant . These results suggest that GIK is involved in a mode of regulation that ensures proper levels of expression of multiple genes during reproductive development ( Figure 7 ) .
Although organ patterning and organogenesis are generally thought to occur independently , evidence has emerged that there is cross-talk between these processes . The homeotic protein AG controls stamen identity partly by activating SPL/NZZ , a gene necessary for specification of male gametophytes [12] . During late stamen development , AG directly controls DAD1 to induce jasmonic acid biosynthesis for stamen maturation [18] . Here we show that another target of AG , GIK , modulates the expression of the auxin response factor ETT through epigenetic modification of the ETT promoter . ETT controls patterning in both the abaxial-adaxial and apical-basal axes of reproductive organs [19]–[21] , [57] . GIK also influences the expression of other key regulators during reproductive development , such as CRC , another abaxial-adaxial polarity-controlling YABBY family gene [25] , [26]; JAG , which is involved in proliferation and differentiation of carpels [28]; and KNU , which is involved in floral meristem determinacy and gametophyte differentiation [30] . Thus , we propose that organ patterning that is mediated by ETT ( and possibly CRC ) and reproductive differentiation that is regulated by KNU and JAG are under partial control of AG , and that GIK acts as a molecular organizer to orchestrate expression of these key regulators for floral reproductive patterning and differentiation ( Figure 7 ) . Ectopic GIK expression in the ag-1 mutant background had minor effects on organ identity and patterning . This does not , however , imply that GIK has no clear function as an AG target . Rather , our data suggest that GIK may modulate and refine spatial and temporal expression of multiple genes downstream of AG . The direct GIK targets , ETT , CRC , JAG , and KNU , are predominantly expressed in reproductive organs , and their expression depends on AG activity to varying degrees . CRC and KNU are directly regulated by AG [13] , [31] . ETT locus is directly bound by SEPALLATA3 , a binding partner of AG [14] . Thus , the effects of ectopic GIK expression were only observed in the wild-type context in which genes downstream of AG are activated . Based on this observation , we conclude that the general role of GIK is to fine-tune the expression of key regulators necessary for patterning and differentiation during reproductive development ( Figure 7 ) . We observed a relatively low penetrance of GIK loss-of-function phenotypes , despite robust phenotypes caused by GIK overexpression . In addition to a possible redundancy , this observation may suggest that GIK does not act as a steadfast controller of gene expression but rather that it fine-tunes the expression of multiple genes through chromatin formation . Furthermore , GIK is expressed in tissues other than flowers , with especially robust expression in roots . Therefore , GIK may have an AG-independent and root-specific function during root growth and development . In agreement with this observation , overexpression of GIK caused root growth inhibition ( Figure S12 ) , even though loss of GIK function did not show clear morphological defects in roots ( unpublished data ) . We showed that ETT is a major target gene for repression by GIK during reproductive development based on the results of a series of genetic and molecular experiments: ( 1 ) GIK overexpression mimics the phenotypes of ett mutants , ( 2 ) GIK and ETT show complementary expression patterns during late reproductive development , ( 3 ) ETT expression is increased in gik mutants , ( 4 ) GIK binds to ETT putative MARs in vivo , and ( 5 ) the putative ETT MARs are important in GIK-regulated ETT expression . GIK-mediated repression of ETT occurred relatively rapidly after GIK induction in floral tissues , and the stable repression of ETT required continuous GIK activity . We also showed that ETT silencing was associated with repressive histone dimethylation of H3K9 in the ETT promoter , especially at the distal putative MAR . It remains unclear whether GIK is directly involved in this histone modification . Because GIK lacks known domains typically found in chromatin-modifying enzymes , GIK may introduce structural changes to the genomic region through MAR binding and may thereby facilitate the binding of chromatin-modifying enzymes to carry out histone modifications . It is also possible that GIK serves as a center for organizing chromatin remodeling complexes in the nuclear matrix to regulate target gene expression . Alternatively , MAR binding by GIK may inhibit the binding of the transcriptional machinery to the proximal promoter , leading to gene silencing associated with dimethylated H3K9 . However , our time-course analysis showed that the dynamic changes in H3K9 dimethylation levels appeared to precede negative regulation of ETT transcription , which does not support the later hypothesis . Dimethylation of H3K9 increased rapidly during the 2 h after GIK induction . A further increase in dimethylation at the 4 h time point corresponded with the steepest downregulation of ETT transcription at the 8 h time point . Conversely , a dynamic reduction of H3K9 dimethylation to a level lower than that seen prior to induction at the 8 h time point was followed by a steady recovery of ETT transcription at the later time points of 16 h and 24 h . Nevertheless , how this dynamic methylation pattern is achieved remains unknown . The mammalian AT-hook protein SATB1 has been shown to mediate gene repression by directly recruiting histone deacetylases [62] . Further studies of proteins that interact with GIK may provide a more detailed account of the mechanism of GIK-mediated repression . ETT has recently been shown to be regulated by trans-acting short interfering RNAs ( siRNAs ) [63]–[65] . Interestingly , ETT expression may be refined by two different molecules , GIK and siRNA , to establish strict spatiotemporal expression boundaries . These events may also partially explain the modest effects of the GIK loss-of-function mutant and of deletion of putative MAR regions in the ETT-promoter reporter construct in the wild-type context . However , it remains to be determined whether GIK and siRNA have separate or overlapping roles in the control of ETT in reproductive development . In mammals , MAR-binding proteins have been implicated in the control of expression of multiple genes . SATB1 in mice contains an AT-hook DNA binding motif and acts as a “gene organizer” to regulate temporal and spatial expression of multiple genes during thymocyte maturation and breast tumor growth and metastasis [40] , [44] . Another SATB1-related MAR-binding protein , SATB2 , represses the expression of several Hox genes during skeletal development and osteoblast differentiation [43] . In agreement with these studies , we show that GIK exhibits similar properties in its regulation of target genes . First , these proteins share the role of a matrix binding protein with an AT-hook DNA binding motif and regulate expression of multiple genes . Second , they are important regulators of various developmental processes: SATB1 in T-cell development , SATB2 in craniofacial patterning and osteoblast differentiation , and GIK in floral reproductive development . Third , most of these proteins execute their effects by modifying chromatin ( SATB1 recruits histone deacetylase , whereas negative regulation by GIK is associated with H3K9 methylation ) . Thus , convergent evolution may have permitted proteins with the same motif to be used for transcriptional coordination in the two kingdoms . Plants and animals are considered to have independently evolved their multicellular developmental processes , but organ or segment identity control in plants and animals starts with transcription factors: HOX genes in animals and MADS genes in plants [66] . Proteins with AT-hook motifs are predominantly present in eukaryotes . The motif is found in some families of HMG proteins that bind to the minor groove of DNA , and the proteins may serve as an anchor for chromatin modifying proteins or may change chromatin architecture [37] , [38] , [62] . Such properties may explain why AT-hook proteins have been used in the evolution of both plant and animal development . In mice , SATB2 controls the expression of the homeotic protein Hoxa2 [43] . In contrast , the homeotic protein AG controls the expression of GIK in Arabidopsis . Thus it is possible that AT-hook motif proteins have been independently incorporated into multicellular developmental processes in animals and plants , but with similar functions of orchestration and fine-tuning of tissue-specific expression of multiple genes .
All plants used in this study are on the Landsberg erecta background and were grown at 22°C under continuous light . DEX treatment was done by submerging inflorescences in a solution containing 10 µM DEX together and 0 . 015% Silwet L-77 for ∼1 min . Transgenic plants were generated by Agrobacterium-mediated infiltration [67] . Plant photographs were taken using a Nikon SMZ 1500 stereoscopic microscope attached to a digital camera ( SIGHT DS-U1 ) . Scanning electron microscope images were taken using a JEOL JSM-6360LV scanning electron microscope . To generate the 35S::GIK and 35S::GIK-GR-6HA constructs , GIK cDNA was cloned into a pMAT137 vector and a composite pGreen vector containing a rat GR hormone binding domain and a 6×HA tag , respectively [12] , [68] . Transgenic plants were selected with kanamycin ( for the pMAT137 construct ) and BASTA ( for the pGreen construct ) for two generations to obtain homozygous lines . 35S::GIK-GR-6HA plants were treated with DEX five times at 1 d intervals for phenotypic observation . More than 90% of DEX-treated flowers showed reproductive defects . A GIK insertion line was obtained from the TRAPPER collection ( http://genetrap . cshl . edu/TrHome . html ) ( NASC stock number , ET14389 ) . The enhancer trap was inserted into the middle of the coding region , 450 bp downstream from the start codon . Homozygous lines were verified by PCR-based genotyping . In total , ∼1% of gik mutant flowers showed reproductive defects . For the rescuing experiment of the gik mutant , a genomic copy of GIK , containing 4 , 660 bp of the 5′ upstream region , 858 bp of the GIK coding region , and 1 , 767 bp of the 3′ region , was cloned into the pDONR221 vector ( Invitrogen ) and later into the pBGW binary vector using gateway cloning [69] for plant transformation . Unexpectedly , we obtained lines showing 5–50-fold higher expression levels of GIK , thus showing the ectopic expression phenotypes . To generate the 35S::GIK-RNAi construct , a C-terminal fragment of the GIK coding region ( GIK-Cter , 410–808 bp ) was amplified using UltraPfu-High-Fidelity DNA polymerase ( Stratagene ) to produce BamHI-GIK-Cter-ClaI and XhoI-GIK-Cter-KpnI fragments . These fragments were cloned into the pKANNIBAL vector [70] . pKANNIBAL-GIK-RNAi was cut by NotI to produce a 35S::GIK-RNAi fragment , which was then cloned into the pMLBART binary vector [71] . GIK-RNAi transgenic plants were selected using BASTA . A few percentages of the examined flowers showed reproductive defects in the T1 and T2 generations . In the T3 generation , the lower ratio of the GIK-RNAi flowers showed reproductive defects . To generate the 35S::GIK2-RNAi construct , an N-terminal fragment of the GIK2 ( AT4g17800; 39–260 bp ) coding region was amplified using UltraPfu-High-Fidelity DNA polymerase to produce BamHI-GIK2-Nter-ClaI and XhoI-GIK2-Nter-KpnI fragments . These fragments were cloned into the pKANNIBAL vector and later into the pMLBART binary vector as described in the cloning process for 35S::GIK-RNAi . A T1 35S::GIK2-RNAi plant was crossed to gik and the GIK2 RNAi gik plants were obtained and confirmed following BASTA selection and PCR genotyping . Full-length GIK cDNA and cDNAs of the conserved N-terminal and AT-hook domains were cloned into the pQE30 vector ( QIAGEN ) to produce 6×His-GIK proteins . Recombinant protein was induced using 1 mM IPTG and purified on a nickel column ( QIAGEN ) under denaturing conditions . Protein was then partially refolded through buffer exchange and concentrated using a Centriprep Centrifugal Filter with an Ultracel YM-10 membrane ( Millipore ) . Purified 6×His-GIK recombinant protein was injected intramuscularly into guinea pigs with Freund's adjuvant . Blood was withdrawn after the fourth and sixth immunizations . Whole blood was processed to obtain polyclonal anti-GIK serum . Approximate 0 . 035 g each of Arabidopsis roots , flowers , and leaves was ground in liquid nitrogen and re-dissolved in 80 µL SDS sample loading buffer ( 0 . 125 M Tris-HCl , pH 6 . 8 , 4% SDS , 10% β-mercaptoethanol , 20% sucrose , 0 . 02% bromophenol blue ) . The samples were boiled for 10 min , and 25 µL of each sample was loaded onto a 12% SDS polyacrylamide gel for electrophoresis . Proteins were transferred onto a PVDF nylon membrane ( Bio-Rad ) and blocked with skim milk . The membrane was then incubated overnight with polyclonal anti-GIK at 4°C , washed with 20 mM Tris-HCl , pH 7 . 5 , 137 mM NaCl , and 0 . 1% [v/v] Tween 20 and further incubated with secondary anti-guinea pig coupled to horseradish peroxidase . Signal was detected using SuperSignal West Dura extended duration substrate ( Pierce ) . A replicate membrane was stained with Coomassie Blue to show protein loading . Arabidopsis seedlings were rinsed with 1× phosphate buffered saline ( PBS ) and fixed with 4% paraformaldehyde in PBS for 1 h . Seedlings were washed three times with PBS and incubated with 4% Driselase ( Sigma ) at 37°C for 30 min . After washing , seedlings were further incubated with PBS containing 10% dimethyl sulfoxide and 3% [v/v] NP-40 for 1 h at room temperature . Seedlings were washed three times with PBS and blocked with 3% bovine serum albumin for 30 min . Seedlings were then incubated overnight with polyclonal anti-GIK or monoclonal anti-trimethylguanosine ( Calbiochem ) . Cy3-conjugated anti-guinea pig and FITC-conjugated anti-mouse were used as secondary antibodies . TOPRO-3 was used as a fluorescent DNA dye . Immunostaining was analyzed with a laser scanning confocal microscope ( Zeiss Meta LSM510 ) . Recombinant proteins GIK-AT ( residues 74-173 ) and GIK-MUT ( residues 74-173; R83K ) were produced in the pQE30 expression vector carried by E . coli M15 cells . ETT MAR probes were generated by cloning SMARTest-predicted MAR sequences [60] in the ETT upstream genomic region into the pCRII vector ( Invitrogen ) . Probe 1 ( −5 , 233 to −5 , 084 bp from translation start site ) and Probe 2 ( −4 , 283 to −4 , 134 from translation start site ) fragments were generated by EcoRI digestion and were end-labeled with a digoxigenin probe synthesis mix ( Roche ) using Klenow fragment ( New England BioLabs ) . South-Western analysis was performed as described [34] with some modifications . Briefly , induced and noninduced bacterial lysates were separated by 10% SDS-PAGE and blotted onto a nitrocellulose membrane ( Bio-Rad ) . The membrane was incubated overnight with 20 ng/mL of digoxigenin-labeled ETT putative MAR probes in DNA binding buffer containing 20 mM Tris-HCl , pH 7 . 4 , 150 mM NaCl , and 20 ng/mL salmon sperm DNA at room temperature , washed , and incubated with anti-dioxigenin coupled with alkaline phosphatase ( Roche ) . Signal was detected using CDP-Star ( Roche ) as a substrate . P1 and P2 probes showed similar binding efficiencies to GIK-AT . The binding result with the P2 probe is shown . Total RNA was isolated from floral bud clusters at stage 10 or younger [18] using the RNeasy plant mini kit ( Qiagen ) and reverse-transcribed using the Superscript III RT-PCR system ( Invitrogen ) . Quantitative real-time PCR assays were performed in triplicate with the 7900HT fast real-time PCR system ( Applied Biosystems ) using the SYBR Green PCR master mix ( Applied Biosystems ) . Statistical analysis was done using paired student's t-test . The ChIP assay was performed as described [18] , [72] with some modifications . Briefly , inflorescences were ground in liquid nitrogen and postfixed with 1% formaldehyde for 10 min . Chromatin was isolated and solubilized by sonication , resulting in an average DNA length of 500 bp . The solubilized chromatin was precleared with salmon sperm DNA-treated protein A- ( for anti-AG , anti-dimethylated H3K9 , and normal rabbit IgG ) or protein G- ( for anti-HA ) agarose beads ( Upstate ) . After centrifugation , the supernatant was incubated overnight with anti-AG ( for AG ChIP experiments ) , anti-HA ( Roche ) ( for GIK ChIP experiments ) , anti-modified histone ( Upstate ) for dimethylated H3K9 , dimethylated H3K4 , acetylated histone H3 , and trimethylated H3K27 ( for histone modification ChIP experiments ) , or normal rabbit IgG ( for both AG and GIK ChIP experiments as a control ) . The DNA-protein complex was precipitated by adding protein A- or protein G-agarose beads , and the purified DNA samples were used for enrichment tests with real-time PCR assays . We measured the ratio between the input DNA before IP and bound DNA after IP for each primer set . The relative enrichment for AG and GIK ChIP experiments was the ratio obtained from: [{ ( Bsp/Isp ) / ( Bctrl/Ictrl ) } of Absp]/[{ ( Bsp/Isp ) / ( Bctrl/Ictrl ) } of control IgG] , where Bsp = amount of bound DNA measured by a specific primer pair; Isp = amount of Input DNA by a specific primer pair; Bctrl = amount of bound DNA by control primer pair ( ACT ) ; Ictrl = amount of Input DNA by control primer pair; and Absp = anti-AG or anti-HA . The control value was set at 1 . 0 . The relative enrichment for the histone modification experiments was the ratio obtained from: [ ( Bsp/Isp ) of X time points]/[ ( Bsp/Isp ) of 0 h] . At least three independent biological replicates of the ChIP assay were performed for the AG ChIP and GIK ChIP experiments . Two independent biological replicates were performed for the histone modification ChIP assay . The real-time PCR assay was done in triplicate for each ChIP assay . One representative data set showing a reproducible trend is shown . Nuclei matrix was isolated as described [59] with some modifications . Briefly , nuclei were isolated using the ChIP method ( see previous section ) without fixation or sonication . The isolated nuclei were washed once with RSB buffer ( 10 mM NaCl , 3 mM MgCl2 , 10 mM Tris-HCl , 0 . 5 mM PMSF , pH 7 . 4 ) and a fraction was kept as a total nuclear control . The remaining sample was digested with 50 U of DNaseI ( Roche ) in RSB containing 0 . 25 M sucrose and 1 mM CaCl2 for 2 h at room temperature . After centrifugation , pellets were resuspended in RSB and an equal volume of high-salt buffer I ( 4 M NaCl , 20 mM EDTA , 20 mM Tris-HCl , pH 7 . 4 ) and incubated for 10 min at 0°C . After centrifugation , the pellets were further extracted twice with high-salt buffer II ( 2 M NaCl , 20 mM EDTA , 20 mM Tris-HCl , pH 7 . 4 , 0 . 25 mg/mL BSA ) . After high-salt extractions , the matrices were washed with RSB buffer containing 0 . 25 M sucrose and 0 . 25 mg/mL BSA and resuspended in the same buffer . The resuspended matrices and total nuclear lysates were used for western analysis . Anti-HA and anti-AG were used to detect GIK-GR-6HA and AG proteins , respectively . To generate the pETT::GUS construct , 8 . 7 kb of ETT upstream genomic sequence was first amplified using UltraPfu-High-Fidelity DNA polymerase with an extension time of 8 min and then cloned into the pDONR221 ( Invitrogen ) to create the entry clone . Similarly , the 4 . 9 kb pETTΔMAR::GUS construct was amplified using UltraPfu-High-Fidelity DNA polymerase with an extension time of 5 min and then cloned into the pENTR directional TOPO cloning vector ( Invitrogen ) . Both clones were sequenced for confirmation . Subsequently , both entry clones were cloned into the pBGWFS7 binary vector [69] using the Gateway cloning method . Transgenic plants with positive GUS reporter expression were crossed with 35S::GIK-GR-6HA plants to obtain pETT::GUS 35S::GIK-GR-6HA and pETTΔMAR::GUS 35S::GIK-GR-6HA double transgenic plants . DEX treatment was performed as described above continuously at 2 d intervals . Whole inflorescences were rinsed and stained to determine GUS activity for GUS expression analysis [73] . Nonradioactive in situ hybridization was performed as described [74] . Full-length ETT cDNA and a 3′ specific region of GIK cDNA were amplified with PCR and cloned into pSK ( Stratagene ) and pCRII vectors , respectively , and used as templates for in vitro transcription . Arabidopsis Genome Initiative locus identifiers of Arabidopsis genes used in this article are as follows: AGAMOUS ( AG , At4g18960 ) , GIANT KILLER ( GIK , At2g35270 ) , ETTIN ( ETT , At2g33860 ) , CRABS CLAW ( CRC , At1g69180 ) , JAGGED ( JAG , At1g68480 ) , KNUCKLES ( KNU , At5g14010 ) , LEUNIG ( LUG , At4g32551 ) , TUBULIN 2 ( TUB , At5g62690 ) , MU-LIKE TRANSPOSASE ( MU , At4g03870 ) , PHOSPHOFRUCTOSE KINASE ( PFK , At4g04040 ) , and LIPASE ( At1g10740 ) .
|
Multicellular development depends on proper expression of thousands of genes . Master regulators , such as homeotic proteins , code for transcription factors in both plants and animals and are thought to act by regulating other genes . Recent genomic studies in the plant Arabidopsis have shown that over 1 , 000 genes are regulated by homeotic proteins that directly control various target genes , including different classes of transcriptional regulators . It is not known , however , how expression of so many genes is coordinated by a single homeotic gene to form functional organs and tissues . Here we identified a transcriptional target of the plant homeotic protein AGAMOUS using bioinformatics analysis and showed that AGAMOUS directly controls GIANT KILLER , a multifunctional chromatin modifier . GIANT KILLER then binds to the upstream regions of multiple genes involved in patterning and differentiation in the AGAMOUS pathway and fine-tunes the expression of these genes . These data therefore provide a possible mechanism by which a homeotic gene coordinates multiple downstream targets in plants .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"biology/plant",
"genetics",
"and",
"gene",
"expression",
"developmental",
"biology/plant",
"growth",
"and",
"development",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
] |
2009
|
AGAMOUS Controls GIANT KILLER, a Multifunctional Chromatin Modifier in Reproductive Organ Patterning and Differentiation
|
The mammalian target of rapamycin ( mTOR ) regulates cell growth and survival by integrating nutrient and hormonal signals . These signaling functions are distributed between at least two distinct mTOR protein complexes: mTORC1 and mTORC2 . mTORC1 is sensitive to the selective inhibitor rapamycin and activated by growth factor stimulation via the canonical phosphoinositide 3-kinase ( PI3K ) →Akt→mTOR pathway . Activated mTORC1 kinase up-regulates protein synthesis by phosphorylating key regulators of mRNA translation . By contrast , mTORC2 is resistant to rapamycin . Genetic studies have suggested that mTORC2 may phosphorylate Akt at S473 , one of two phosphorylation sites required for Akt activation; this has been controversial , in part because RNA interference and gene knockouts produce distinct Akt phospho-isoforms . The central role of mTOR in controlling key cellular growth and survival pathways has sparked interest in discovering mTOR inhibitors that bind to the ATP site and therefore target both mTORC2 and mTORC1 . We investigated mTOR signaling in cells and animals with two novel and specific mTOR kinase domain inhibitors ( TORKinibs ) . Unlike rapamycin , these TORKinibs ( PP242 and PP30 ) inhibit mTORC2 , and we use them to show that pharmacological inhibition of mTOR blocks the phosphorylation of Akt at S473 and prevents its full activation . Furthermore , we show that TORKinibs inhibit proliferation of primary cells more completely than rapamycin . Surprisingly , we find that mTORC2 is not the basis for this enhanced activity , and we show that the TORKinib PP242 is a more effective mTORC1 inhibitor than rapamycin . Importantly , at the molecular level , PP242 inhibits cap-dependent translation under conditions in which rapamycin has no effect . Our findings identify new functional features of mTORC1 that are resistant to rapamycin but are effectively targeted by TORKinibs . These potent new pharmacological agents complement rapamycin in the study of mTOR and its role in normal physiology and human disease .
The mammalian target of rapamycin ( mTOR ) is a serine-threonine kinase related to the lipid kinases of the phosphoinositide 3-kinase ( PI3K ) family . mTOR exists in two complexes , mTORC1 [1 , 2] and mTORC2 [3 , 4] , which are differentially regulated , have distinct substrate specificities , and are differentially sensitive to rapamycin . mTORC1 integrates signals from growth factor receptors with cellular nutritional status and controls the level of cap-dependent mRNA translation by modulating the activity of key translational components such as the cap-binding protein and oncogene eIF4E [5] . mTORC2 is insensitive to rapamycin , and selective inhibitors of this complex have not been described . Partly because acute pharmacological inhibition of mTORC2 has not been possible , the functions of mTORC2 are less well understood than those of mTORC1 . mTORC2 is thought to modulate growth factor signaling by phosphorylating the C-terminal hydrophobic motif of some AGC kinases such as Akt [3 , 6] and SGK [7] although other kinases , including DNA-PK and Ilk , have also been implicated in Akt hydrophobic motif phosphorylation [8–11] . Growth factor stimulation of PI3K causes activation of Akt by phosphorylation at two key sites: the activation loop ( T308 ) and the C-terminal hydrophobic motif ( S473 ) . Active Akt promotes cell survival in many ways , including suppressing apoptosis , promoting glucose uptake , and modifying cellular metabolism [12]; consequently , there is significant interest in identifying the kinase ( s ) responsible for each activating phosphorylation , the relationship between these phosphorylation sites , and the role of differential Akt phosphorylation on Akt substrate phosphorylation . Of the two phosphorylation sites on Akt , activation loop phosphorylation at T308 , which is mediated by PDK1 , is indispensable for kinase activity , whereas hydrophobic motif phosphorylation at S473 enhances Akt kinase activity by approximately 5-fold [13] . The disruption of mTORC2 by different genetic and pharmacological approaches has variable effects on Akt phosphorylation . Targeting mTORC2 by RNA interference ( RNAi ) [6 , 14] , homologous recombination [15–17] , or long-term rapamycin treatment [18] results in loss of Akt hydrophobic motif phosphorylation ( S473 ) , strongly implicating mTORC2 as the kinase responsible for phosphorylation of this site . RNAi targeting mTORC2 and long-term rapamycin result in loss of Akt phosphorylation on its activation loop ( T308 ) , but this phosphorylation remains intact in mouse embryonic fibroblasts ( MEFs ) lacking the critical mTORC2 component SIN1 . It cannot be inferred from this genetic data whether acute pharmacological inhibition of mTORC2 would block the phosphorylation of Akt only at S473 , resulting in partial Akt deactivation , or also disrupt phosphorylation at T308 , resulting complete Akt inhibition . Several small molecules have been identified that directly inhibit mTOR by targeting the ATP binding site; these include LY294002 , PI-103 , and NVP-BEZ235 [19–22] . These molecules were originally discovered as inhibitors of PI3Ks and later shown to also target mTOR . Because all of these molecules inhibit PI3Ks and mTOR with similar potency , they cannot be used to selectively inhibit mTOR or PI3Ks in cells . Indeed , because mTORC1 and mTORC2 function downstream of PI3Ks in most settings , it is unclear to what extent the ability of these molecules to block the activation of signaling proteins such as Akt reflects PI3K versus mTOR inhibition . It is possible that some of the functions attributed to PI3Ks using the classical inhibitor LY294002 are a consequence of mTOR inhibition [19 , 23] , but it is has not been possible address this , because small molecules that inhibit mTOR without inhibiting PI3Ks have not been available . We recently reported the synthesis of pyrazolopyrimidines that inhibit members of the PI3K family , including mTOR [24] . Two of these molecules , PP242 and PP30 , are the first potent , selective , and ATP-competitive inhibitors of mTOR . Unlike rapamycin , these molecules inhibit both mTORC1 and mTORC2 , and , unlike PI3K family inhibitors such as LY294002 , these molecules inhibit mTOR with a high degree of selectivity relative to PI3Ks and protein kinases . To distinguish these molecules from the allosteric mTORC1 inhibitor rapamycin , we are calling them “TORKinibs” for TOR kinase domain inhibitors . The dual role of mTOR within the PI3K→Akt→mTOR pathway as both an upstream activator of Akt and the downstream effector of pathway activity on cell growth and proliferation has excited interest in active-site inhibitors of mTOR [25–30] . We describe here the biological activity of these molecules . Another small-molecule ATP-competitive mTOR inhibitor called Torin1 was reported while our manuscript was in the process of publication [56] .
PP242 and PP30 inhibit mTOR in vitro with half-maximal inhibitory concentrations ( IC50 values ) of 8 nM and 80 nM , respectively . As expected for active-site inhibitors , PP242 and PP30 inhibit mTOR in both mTORC1 and mTORC2 ( Table S1 ) . Both compounds are selective within the PI3K family , inhibiting other PI3Ks only at substantially higher concentrations ( Figure 1 ) . Testing of PP242 against 219 purified protein kinases at a concentration 100-fold higher than its mTOR IC50 value revealed exceptional selectivity with respect to the protein kinome; most protein kinases were unaffected by this drug , and only four—PKC-alpha , PKC-beta , RET , and JAK2 ( V617F ) —were inhibited more than 80% [24] . We determined IC50 values for PP242 against these kinases in vitro using purified proteins . In these assays , PP242 was relatively inactive against PKC-beta , RET , or JAK2 but inhibited PKC-alpha with an in vitro IC50 of 50 nM ( Figure 1 ) . Importantly , PP30 showed no activity against PKC-alpha or PKC-beta in the same assay ( Figure 1 ) . These data indicate that PP242 is a highly selective inhibitor of mTOR and that PP30 can be used to confirm that the effects of PP242 are due to inhibition of mTOR and not PKC-alpha . The availability of a second structurally dissimilar mTOR inhibitor—PP30—provides additional control for unanticipated off-targets of PP242 . We characterized the effect of PP242 on the PI3K→Akt→mTOR pathway . PP242 and PP30 both inhibited insulin-stimulated phosphorylation of Akt at S473 , confirming that mTOR kinase activity is required for hydrophobic motif phosphorylation ( Figure 2A ) . The inhibition of mTOR by PP242 and PP30 also resulted in loss of Akt phosphorylation at T308 , but significantly higher doses of PP242 and PP30 were required to inhibit T308 as compared with S473 ( Figure 2A and 2B ) . PP242 inhibited S473-P and T308-P at both early and late time points after insulin stimulation , indicating that the differential sensitivity of these sites to PP242 does not reflect differing kinetics of phosphorylation ( Figure S1 ) . By comparison , the PI3K inhibitor PIK-90 , which does not inhibit mTOR , inhibited the phosphorylation of both Akt sites equipotently ( Figure 2B ) , as observed previously [21] . We sought to confirm that the loss of T308-P caused by PP242 and PP30 results from inhibition of mTOR-mediated phosphorylation of S473 , rather than from inhibition of an off-target kinase , or from an effect of mTOR inhibition unrelated to S473-P . To do this , we examined the effect of PP242 on T308 phosphorylation in two situations in which Akt could not be phosphorylated on S473 . First , we overexpressed S473A mutant Akt and stimulated these cells with insulin ( Figure 3A ) . S473A Akt was phosphorylated on T308 to a similar level as wild-type , yet in contrast to the wild-type , T308-P on S473A Akt was not inhibited by PP242 . The lack of effect of PP242 on S473A Akt confirms that PP242 inhibition of pT308 requires S473 and also that PP242 does not inhibit PDK1 in cells , as was suggested by direct testing of PDK1 in vitro ( Figure 1 ) . As a further test of the specificity of PP242 and the requirement for functional S473 phosphorylation in order for PP242 to inhibit T308-P , we examined the effect of PP242 on the phosphorylation of Akt in primary MEFs from embryos that lack SIN1 [16] ( Figure 3B ) . SIN1 is a component of mTORC2 , and knockout of SIN1 compromises the physical integrity of mTORC2 leading to a complete loss of Akt phosphorylation at S473 without affecting its phosphorylation at T308 . Consistent with our results from L6 cells , PP242 inhibited the phosphorylation of Akt at both S473 and T308 in wild-type MEFs . By contrast , PP242 had no effect on the phosphorylation of T308 in SIN1−/− MEFs that lack mTORC2 . Furthermore , PP242 had no effect on the constitutive phosphorylation of the turn motif of Akt at T450 [16 , 31] . As a further comparison , we examined the effect of long-term rapamycin , which is known to block the assembly of mTORC2 is some cell lines [18] . Similar to PP242 , long-term rapamycin treatment of wild-type MEFs inhibited S473-P and reduced the phosphorylation of T308-P , as was seen previously [18] . Importantly , the PI3K inhibitor PIK-90 and the PDK1 inhibitor BX-795 [32] blocked phosphorylation of T308 in SIN1−/− MEFs , indicating that the failure of PP242 to block T308 in SIN1−/− MEFs does not reflect a general resistance of T308 to dephosphorylation in cells that lack mTORC2 . From these data , we conclude that PP242′s effect on T308-P is dependent on its inhibition of Akt phosphorylation by mTOR at S473 . It remains unclear why mTORC2 knockout cells , but not cells treated with RNAi or pharmacological inhibitors of mTORC2 , are able to retain T308 phosphorylation in the absence of phosphorylation at S473 . However , there are a growing number of examples in which genetic deletion of a kinase results in compensatory changes that mask relevant phenotypes observed with the corresponding small molecule inhibitor [33] . Akt requires phosphorylation at both S473 and T308 for full biochemical activity in vitro [13] , but it is unclear whether all of the cellular functions of Akt require it to be dually phosphorylated . Singly phosphorylated ( T308-P ) Akt from SIN1−/− MEFs is competent to phosphorylate the cytoplasmic Akt substrates GSK3 and TSC2 , but not the nuclear target FoxO [16] . Because low concentrations of PP242 inhibit the phosphorylation of S473 and higher concentrations partially inhibit T308-P in addition to S473-P , we used PP242 to examine whether some substrates of Akt are especially sensitive to loss of S473-P ( Figure 4 ) . We compared PP242 to the PI3K inhibitor PIK-90 and the allosteric Akt inhibitor Akti-1/2 [34] , which inhibit the phosphorylation of Akt at both sites . In contrast to PIK-90 and Akti-1/2 , which completely inhibited the phosphorylation of Akt and its direct substrates , PP242 only partially inhibited the phosphorylation of cytoplasmic and nuclear substrates of Akt . This suggests that phosphorylation of the Akt substrates we examined is only modestly sensitive to loss of S473-P . A caveat of comparing Akt substrates in Sin1−/− MEFs with PP242-treated cells is the different turn motif ( T450-P ) status in these two conditions ( Figure 3B ) . In contrast to Akt , which maintains T308-P , SGK activity is completely inhibited by genetic disruption of mTORC2 [7] . Because SGK can phosphorylate FoxO and its activity is completely inhibited by disruption of mTORC2 , it was suggested that the loss of FoxO phosphorylation in SIN1−/− MEFs indicates that FoxO is primarily phosphorylated by SGK rather than Akt [7] . Because Akti-1/2 does not inhibit SGK [34] but inhibits FoxO1/O3a phosphorylation at T24/T32 in L6 myotubes ( Figure 4 ) , our data suggests that the major kinase for T24/T32 of FoxO1/O3a in L6 myotubes is Akt and not SGK . TORC2 is required for the generation of a polarized actin cytoskeleton in yeast [35] . Previous analysis of mTORC2 function using RNAi revealed a role for mTORC2 in the control of the actin cytoskeleton [3 , 4] , yet these findings were not confirmed in primary MEFs lacking mTORC2 [15 , 17] . We examined actin stress fibers in NIH 3T3 cells ( Figure 5 ) and in primary MEFs ( unpublished data ) treated with PP242 . After 8 h of treatment with PP242 , we found no obvious effect on the morphology or abundance of actin stress fibers ( Figure 5 ) , suggesting that mTORC2 activity is not required for the maintenance of actin stress fibers in these cells . That PP242 didn't obviously affect the morphology or abundance of actin stress fibers , does not rule out a role for mTOR in the control of the actin cytoskeleton , but it does show that pharmacological inhibition of mTORC2 does not affect the obvious changes in actin structure seen with RNAi . We next measured the effect of dual mTORC1/mTORC2 inhibition by PP242 on the proliferation of primary MEFs ( Figure 5B ) . For this analysis , we compared PP242 to selective mTORC1 inhibition by rapamycin . Rapamycin was tested at concentrations above its mTOR IC50 , and at all concentrations tested , it inhibited growth to the same extent . By contrast , PP242 had a dose-dependent effect on proliferation and at higher doses was much more effective than rapamycin at blocking cell proliferation . The ability of PP242 to block cell proliferation more efficiently than rapamycin could be a result of its ability to inhibit mTORC1 and mTORC2 , because rapamycin can only inhibit mTORC1 . To test this possibility , we measured the effects of both compounds on the proliferation of SIN1−/− MEFs , which lack mTORC2 . In SIN1−/− MEFs , rapamycin was also less effective at blocking cell proliferation than PP242 . That PP242 and rapamycin exhibit very different anti-proliferative effects in SIN1−/− MEFs suggests that the two compounds differentially affect mTORC1 . mTORC1 regulates protein synthesis by phosphorylating the hydrophobic motif of p70S6-Kinase ( S6K ) at T389 and the eIF4E-binding-protein , 4EBP1 , at multiple sites . Our proliferation experiments suggest that rapamycin and PP242 have distinct effects on mTORC1 . We compared the effects of acute treatment with rapamycin and PP242 on S6K , ribosomal protein S6 ( S6 ) , and 4EBP1 phosphorylation ( Figure 6A ) to see if these inhibitors differentially affect the phosphorylation of these canonical substrates of mTORC1 . Both rapamycin and PP242 inhibited the phosphorylation of S6K and its substrate S6 , and neither rapamycin nor PP242 affected the phosphorylation of 4EBP1 on T70 ( Figure S2A ) . In contrast , PP242 fully inhibited the phosphorylation of 4EBP1 at T36/45 and S65 , whereas rapamycin only had a modest affect on these same phosphorylations . Treatment of cells with PP30 was also effective at reducing the phosphorylation of 4EBP1 at T36/45 ( Figure S3 ) , indicating that the block of T36/45 phosphorylation by PP242 is due to its inhibition of mTOR and not PKC-alpha . PIK-90 did not reduce the phosphorylation of 4EBP1 at T36/45 , demonstrating that inhibition of PI3K and Akt activation alone is not sufficient to block the phosphorylation of 4EBP1 at T36/45 ( Figure S3 ) . The enhanced dephosphorylation of 4EBP1 caused by PP242 as compared with rapamycin could be due to incomplete inhibition of mTORC1 by rapamycin or involvement of mTORC2 in the phosphorylation of 4EBP1 . To examine these alternatives , we analyzed the effect of PP242 and rapamycin on the phosphorylation of 4EBP1 in SIN1−/− MEFs that lack mTORC2 ( Figure 6B ) . SIN1−/− MEFS showed higher levels of p4EBP1 , suggesting that due to the lack of mTORC2 , these cells have more mTORC1 activity , although stronger S6K phosphorylation in wild-type cells contradicts this simple interpretation . Despite an increase in p4EBP1 in SIN1−/− compared with wild-type MEFs , shorter exposures of the p4EBP1 blots ( Figure S2B ) show that PP242 inhibits p4EBP1 with the same potency in both cells . The fuller inhibition of p4EBP1 by PP242 than by rapamycin in wild-type and SIN1−/− MEFs indicates that the presence of mTORC2 is not required for rapamycin and PP242 to have distinct effects on 4EBP1 phosphorylation , and suggests that PP242 is a more complete inhibitor of mTORC1 than rapamycin . While the precise role of S6K in translation control is still poorly understood , it is known that the hypophosphorylated 4EBP1 protein acts as negative regulator of the major cap-binding protein eIF4E . We directly assessed the effect of PP242 on cap-dependent translation downstream of mTOR activation . The phosphorylation of 4EBP1 by mTOR in response to growth factor and nutrient status causes it to dissociate from eIF4E allowing eIF4G and associated factors to bind to the 5' cap , recruit the 40S subunit of the ribosome , and scan the mRNA for the start codon to initiate translation . The phosphorylation of 4EBP1 by mTOR is complicated in that it occurs at multiple sites , and not all sites are equally effective at causing dissociation of 4EBP1 from eIF4E [36] . Furthermore , a hierarchy is thought to exist whereby the N-terminal threonine phosphorylations at 36/45 precede and are required for the C-terminal phosphorylations at S65 and T70 [37 , 38] . Phosphorylation at S65 causes the greatest decrease in affinity of 4EBP1 for eIF4E [39 , 40] , and S65 is probably the most important site in cells for dissociation of 4EBP1 from eIF4E [41] , but other sites are also important [36 , 42] . We examined the effect of PP242 on the active eIF4E initiation complex of translation by using a cap-binding assay . eIF4E binds tightly to beads coated with the cap analogue 7-methyl GTP ( m7GTP ) , allowing proteins bound to eIF4E to be examined . Rapamycin caused partial inhibition of the insulin-stimulated release of 4EPB1 from eIF4E ( Figure 7A ) , consistent with its partial inhibition of S65 phosphorylation ( Figure 6A ) . The rapamycin-induced retention of 4EBP1 was accompanied by a loss of recovery of eIF4G , because the binding of 4EBP1 and eIF4G to eIF4E are mutually exclusive . In contrast , treatment with PP242 caused a much larger retention of 4EBP1 , raising the retention of 4EBP1 above the level seen in unstimulated serum-starved cells , which are known to have low levels of protein translation [43] . Translation initiation depending on eIF4E activity is the rate-limiting step in cap-dependent protein translation [44] . PP242 caused a higher level of binding between 4EBP1 and eIF4E than rapamycin ( Figure 6A ) , suggesting that cap-dependent translation will be more highly suppressed by PP242 than by rapamycin . To quantify the efficiency of cap-dependent translation in the presence of PP242 and rapamycin , we used the well-established bicistronic reporter assay where translation initiation of the first cistron is dependent on the 5′ cap , whereas initiation of the second cistron depends on a viral internal ribosome entry site ( IRES ) that bypasses the need for cap-binding proteins such as eIF4E [45] . PP242 caused a significant decrease in cap-dependent , but not IRES-dependent ( Figure S4 ) , translation , whereas rapamycin did not have a statistically significant effect on cap-dependent translation ( Figure 7B ) , consistent with the modest effect of rapamycin on 4EBP1 phosphorylation ( Figure 6A ) . Based on this assay , inhibition of mTOR and p4EBP1 reduces cap-dependent translation by about 30% , suggesting that cap-dependent translation is only partially inhibited by hypophosphorylated 4EBP1 . The majority of protein synthesis is thought to be cap-dependent [44] , and consistent with this we find that PP242 also reduces total protein synthesis by about 30% , whereas rapamycin does not have a significant effect ( Figure 7C and 7D ) . Mouse knock-outs of mTORC1 or mTORC2 result in embryonic lethality and thus it has been difficult to examine the effects of loss of mTOR in animals . To begin to explore the tissue specific roles of mTORC1 and mTORC2 and confirm the pathway analysis from cell culture experiments , we treated mice with PP242 and rapamycin and examined the acute effect of these drugs on insulin signaling in fat , skeletal muscle , and liver tissue ( Figure 8 ) . In fat and liver , PP242 was able to completely inhibit the phosphorylation of Akt at S473 and T308 , consistent with its effect on these phosphorylation sites observed in cell culture . Surprisingly , PP242 was only partially able to inhibit the phosphorylation of Akt in skeletal muscle and was more effective at inhibiting the phosphorylation of T308 than S473 , despite it's ability to fully inhibit the phosphorylation of 4EBP1 and S6 . These results will be confirmed by in vivo dose-response experiments , but , consistent with the partial effect of PP242 on pAkt in skeletal muscle , a muscle-specific knockout of the integral mTORC2 component rictor resulted in only a partial loss of Akt phosphorylation at S473 [46] . These results suggest that a kinase other than mTOR , such as DNA-PK [8 , 9] , may contribute to phosphorylation of Akt in muscle . Rapamycin often stimulates the phosphorylation of Akt [47 , 48] , probably by relieving feedback inhibition from S6K to the insulin receptor substrate 1 ( IRS1 ) [49] , a key signaling molecule that links activation of the insulin receptor to PI3K activation . In all tissues examined , and especially in fat and muscle , acute rapamycin treatment activated the phosphorylation of Akt at S473 and T308 ( Figure 8 ) . In contrast to rapamycin , by inhibiting both mTORC2 and mTORC1 , PP242 suppresses rather than enhances Akt activation . As was seen in cell culture , rapamycin and PP242 also differentially affect the mTORC1 substrates S6K and 4EBP1 in vivo . S6 phosphorylation was fully inhibited by rapamycin and PP242 in all tissues examined . While PP242 was effective at blocking the phosphorylation of 4EBP1 on both T36/45 and S65 in all tissues examined , rapamycin did not block 4EBP1 phosphorylation as completely as PP242 . Further experiments will be required to identify the mechanism by which 4EBP1 phosphorylation is partially resistant to rapamycin .
Rapamycin has been a powerful pharmacological tool allowing the discovery of mTOR's role in the control of protein synthesis . Since the discovery of a rapamycin-insensitive mTOR complex , there has been a significant effort to develop pharmacological tools for studying this complex . We have used two structurally distinct compounds to pharmacologically dissect the effects of mTOR kinase inhibition toward mTORC1 and mTORC2 activity . We have shown through the use of these inhibitors that the inhibition of mTOR kinase activity is sufficient to prevent the phosphorylation of Akt at S473 , providing further evidence that mTORC2 is the kinase responsible for Akt hydrophobic motif phosphorylation upon insulin stimulation . We also find that phosphorylation at T308 is linked to phosphorylation at S473 , as had been observed in experiments where mTORC2 was disabled by RNAi and long-term rapamycin , but not homologous recombination . Surprisingly however , inhibition of mTORC2 does not result in a complete block of Akt signaling , as T308P is partially maintained and Akt substrate phosphorylation is only modestly affected when S473 is not phosphorylated . Despite its modest effect on Akt substrate phosphorylation , PP242 was a strikingly more effective anti-proliferative agent than rapamycin . These results were reproduced even in cells lacking mTORC2 ( SIN1−/− ) , suggesting that downstream mTORC1 substrates might be responsible for PP242′s strong anti-proliferative effects . Interestingly , we observe that phosphorylation of the mTORC1 substrate 4EBP1 is partially resistant to rapamycin treatment at concentrations that fully inhibit S6K , whereas PP242 completely inhibits both S6K and 4EBP1 . Because rapamycin can only partially inhibit the phosphorylation of 4EBP1 , but it can fully in inhibit the phosphorylation of S6K , rapamycin appears to be a substrate-selective inhibitor of mTORC1 . Consistent with this finding , experiments with purified proteins have shown that rapamycin/FKBP12 only partially inhibits the in vitro phosphorylation of 4EBP1 at Ser 65 by mTOR but can fully inhibit the in vitro phosphorylation of S6K [50] . By contrast , LY294002 , a direct inhibitor of many PI3K family members including mTOR , was equally effective at inhibiting the phosphorylation of S6K and 4EBP1 by mTOR in vitro [50] and in cells [23] , although this finding is complicated by LY294002′s inhibition of multiple lipid and protein kinases [51] including PIM , a kinase potentially upstream of 4EBP1 phosphorylation [52 , 53] . These results argue that PP242 , in addition to being useful for investigating mTORC2 , can reveal rapamycin-resistant components of mTORC1 function . Indeed , proliferation of SIN1−/− MEFs is more sensitive to PP242 than rapamycin ( Figure 5B ) , suggesting that rapamycin-resistant functions of mTORC1 , including the aspects of translation initiation highlighted in Figure 7 , are key to the anti-proliferative effects of PP242 . Furthermore , our findings suggest that the inhibition of translational control and the anti-proliferative effects of PP242 require inhibition of 4EBP1 phosphorylation and eIF4E activity . Using TORKinibs to acutely inhibit mTOR has surprisingly led to the identification of outputs from mTORC1 that are rapamycin-resistant . These observations should motivate further studies aimed at understanding how rapamycin is able to selectively affect different outputs downstream of mTORC1 . As active site inhibitors of mTOR join rapamycin and its analogs in the clinic [22 , 27 , 30] , it will be important to understand the distinct effects of these pharmacological agents on cellular and organismal physiology and to evaluate their efficacy in the treatment of disease and cancer caused by hyperactivation of the PI3K→Akt→TOR pathway .
Mice were handled in accordance with protocols approved by the committee for animal research at the University of California San Francisco , United States of America . Cells were grown in DMEM supplemented with 10% FBS , glutamine , and penicillin/streptomycin . Confluent L6 myoblasts were differentiated into myotubes by culturing them for 5 d in medium containing 2% FBS . L6 myotubes were maintained in medium containing 2% FBS until use . Primary wild-type MEFs used in Figure 7 were isolated at embryonic day 13 . 5 as previously described [54] . Primary SIN1−/− MEFs and matching wild-type controls were provided by B . Su and isolated as previously described [16] . Except where indicated otherwise , cells were serum starved overnight and incubated with inhibitors or 0 . 1% DMSO for 30 min prior to stimulation with 100 nM insulin for 10 min . All inhibitors were either synthesized as previously described [21 , 24 , 55] or were from Calbiochem ( rapamycin and Akti-1/2 ) . Cells were lysed by scraping into ice cold lysis buffer followed by brief sonication . Lysates were cleared by centrifugation , resolved by SDS-PAGE , transferred to nitrocellulose , and immunoblotted with antibodies from Cell Signaling Technology . Unless otherwise indicated , cells were lysed in 300 mM NaCl , 50 mM Tris pH 7 . 5 , 5 mM EDTA , 1% Triton X-100 , 0 . 02% NaN3 , 20 nM microcystin ( Calbiochem ) , Sigma phosphatase inhibitor cocktails 1 and 2 , Roche protease inhibitor cocktail , and 2 mM PMSF . For Figures 6A and 7A , and Figure S2A , cells were lysed in cap lysis buffer ( 140 mM KCl , 10 mM Tris pH 7 . 5 , 1 mM EDTA , 4 mM MgCl2 , 1 mM DTT , 1% NP-40 , 20 nM microcystin , Sigma phosphatase inhibitor cocktails 1 and 2 , Roche protease inhibitor cocktail without EDTA and 2 mM PMSF ) . L6 myotubes from one well of a six-well plate were lysed in 300 μl of cap lysis buffer as described above . 50 μl of detergent-free cap lysis buffer and 20 μl of pre-washed cap beads were added to 150 μl of cleared lysate and incubated at 4 °C overnight with tumbling . The beads were washed twice with 400 μl of cap wash buffer ( cap lysis buffer with 0 . 5% NP-40 instead of 1% NP-40 ) and twice with 500 μl of PBS . The beads were boiled in SDS-PAGE sample buffer and the retained proteins analyzed by Western blot . All antibodies were from Cell Signaling Technologies except for the anti-eIF4E antibody , which was from BD Biosciences . Phosphorylation of histone H1 ( 4 μM ) by PKC was assayed in a buffer containing 200 ng/ml recombinant kinase , 25 mM HEPES pH 7 . 5 , 10 mM MgCl2 , 5 mM ß-glycerol phosphate , 0 . 05 mg/ml phosphatidylserine , 0 . 03% Triton X-100 , 0 . 5 mg/ml BSA , 2 . 5 mM DTT , 100 μM CaCl2 , 1 μM PMA , 10 μM ATP , and 15 μCi/ml of γ-32P-ATP . Inhibitors were tested in a four-fold dilution series from 10 μM to 600 pM , and four measurements were made at each concentration . The kinase reaction was terminated by spotting onto nitrocellulose , which was washed 5 times with 1 M NaCl/1% phosphoric acid . The radioactivity remaining on the nitrocellulose sheet was quantified by phosphorimaging , and IC50 values were determined by fitting the data to a sigmoidal dose-response curve using the Prism software package . PDK1 , mTORC1 , and mTORC2 were assayed as previously described [21] . L6 myotubes were grown and differentiated in 96-well plates . The outside wells of the plate were not used for the experiment , but were kept filled with media . Following stimulation , cells were fixed for 15 min with 4% formaldehyde in PBS with Ca++ and Mg++ . The cells were washed three times with PBS and the blocked and permeabilized with 5% goat serum in PBS with 0 . 3% Triton X-100 ( PBS-GS-TX ) . Primary antibodies to S473 ( Cell Signaling #4060 ) and T308 ( Cell Signaling #2965 ) were added at 1:1000 and 1:500 , respectively , in PBS-GS-TX , and the plates were incubated at 4 °C overnight . The plates were then washed three times with PBS , and goat anti-rabbit secondary antibody ( Pierce Biotechnology ) was added at 0 . 01 μg/ml in PBS-GS-TX . After 1 h at room temperature , plates were washed three times with PBS . ELISA chemiluminescent reagent ( Femto , Pierce Biotechnology ) was added to each well and after 1 min , the plate was read in a luminescence plate reader using a 100-ms integration time . The pAkt signal from pT308 and pS473 was normalized to control wells , so that 0 represents the level of pAkt in serum starved cells and 1 represents the level upon insulin stimulation . EC50 values were determined by fitting the data to a sigmoidal dose-response curve using the Prism software package . The significance of differences between EC50 values was evaluated using the F test . Akt was transfected into HEK293 cells using Lipofectamine 2000 according the manufacturers protocol . Two days after transfection , cells were serum starved overnight and the next day they were treated with inhibitors and processed for western blotting as described above . NIH 3T3 cells were plated on poly-lysine coated coverslips at 30% confluence the day before the experiment . Following treatment with PP242 or 0 . 1% DMSO for 8 h in 10% serum growth medium , the actin cytoskeleton was stained as previously described [24] . Primary MEFs were transfected with a bicistronic reporter [54] containing a viral IRES using Lipofectamine 2000 according to the manufacturers protocol . At 2 d post transfection , cells were treated overnight with compounds as indicated or starved of serum . The next day , Renilla and Firefly luciferase activity were measured using the Dual-Luciferase kit ( Promega ) . Differences in the ratio of Renilla to Firefly luciferase signals were analyzed for statistical significance by one-way ANOVA with Tukey's post test using the Prism software package . Primary MEFs grown to 70% confluence in six-well plates were incubated overnight in either 10% Serum ( Steady State ) , kinase inhibitors in 10% serum , or 0 . 1% serum ( starved ) . Cells were then washed once with DMEM lacking cysteine and methionine ( DMEM-noS ) , and the medium was replaced with DMEM-noS including dialyzed serum and kinase inhibitors as indicated . After incubation for 1 h , 50 μCi of Expre35S35S ( NEN ) was added to each well and the cells were labeled for 4 h . Cells were washed once with ice-cold PBS , and lysed as described above for Western blotting . Following separation by SDS-PAGE , and transfer to nitrocellulose , 35S-labeled proteins were visualized by autoradiography with film . For quantitation , the membrane was exposed to a phosphorimager screen and the resulting image was quantified in ImageJ . Differences in 35S incorporation were analyzed for statistical significance by one-way ANOVA with Tukey's post test using the Prism software package . Drugs were prepared in 100 μl of vehicle containing 20% DMSO , 40% PEG-400 , and 40% saline . Six-wk-old male C57BL/6 mice were fasted overnight prior to drug treatment . PP242 ( 0 . 4 mg ) , rapamycin ( 0 . 1 mg ) , or vehicle alone was injected IP . After 30 min for the rapamycin-treated mouse or 10 min for the PP242 and vehicle-treated mice , 250 mU of insulin in 100 μl of saline was injected IP . 15 min after the insulin injection , the mice were killed by CO2 asphyxiation followed by cervical dislocation . Tissues were harvested and frozen on liquid nitrogen in 200 μl of cap lysis buffer . The frozen tissue was thawed on ice , manually disrupted with a mortar and pestle , and then further processed with a micro tissue-homogenizer ( Fisher PowerGen 125 with Omni-Tip probe ) . Protein concentration of the cleared lysate was measured by Bradford assay and 5–10 μg of protein was analyzed by Western blot as described above . Wild-type and SIN1−/− MEFs were plated in 96-well plates at approximately 30% confluence and left overnight to adhere . The following day cells were treated with PP242 , rapamycin , or vehicle ( 0 . 1% DMSO ) . After 72 h of treatment , 10 μl of 440 μM resazurin sodium salt ( Sigma ) was added to each well , and after 18 h , the florescence intensity in each well was measured using a top-reading florescent plate reader with excitation at 530 nm and emission at 590 nm .
|
Growth factor pathways are required for normal development but are often inappropriately activated in many cancers . One growth-factor–sensitive pathway of increasing interest to cancer researchers relies on the mammalian target of rapamycin ( mTOR ) , a kinase that ( like all kinases ) delivers phosphate groups from ATP to amino acid residues of downstream proteins . TOR proteins were first discovered in yeast as the cellular targets of rapamycin , a small , naturally occurring molecule derived from bacteria that is widely used as an immunosuppressant and more recently in some cancer therapies . The study of TOR proteins has relied heavily on the use of rapamycin , but rapamycin does not directly inhibit TOR kinase activity; rather , rapamycin influences TOR's enzymatic activities by binding to a domain far from the kinase's active site . Some mTOR functions are resistant to rapamycin , as a result of the kinase activity of one kind of multiprotein complex , the mTOR complex 2 ( mTORC2 ) , whereas rapamycin-sensitive functions of mTOR are due to the mTOR complex 1 ( mTORC1 ) . We have developed new inhibitors of mTOR that bind to the ATP-binding site of mTOR and inhibit the catalytic activity of both mTORC1 and mTORC2 without inhibiting other kinases . Unexpectedly , these inhibitors had profound effects on protein synthesis and cell proliferation due to their inhibition of mTORC1 rather than mTORC2 . We found that the phosphorylation of a protein that controls protein synthesis , the mTORC1 substrate 4E binding protein ( 4EBP ) is partially resistant to rapamycin but fully inhibited by our new inhibitors . The finding that 4EBP phosphorylation is resistant to rapamycin suggests that active-site inhibitors may be more effective than rapamycin in the treatment of cancer and may explain why rapamycin is so well tolerated when taken for immunosuppression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"cell",
"biology",
"pharmacology",
"chemical",
"biology",
"biophysics"
] |
2009
|
Active-Site Inhibitors of mTOR Target Rapamycin-Resistant Outputs of mTORC1 and mTORC2
|
Current strategies to improve graft outcome following kidney transplantation consider information at the human leukocyte antigen ( HLA ) loci . Cell surface antigens , in addition to HLA , may serve as the stimuli as well as the targets for the anti-allograft immune response and influence long-term graft outcomes . We therefore performed exome sequencing of DNA from kidney graft recipients and their living donors and estimated all possible cell surface antigens mismatches for a given donor/recipient pair by computing the number of amino acid mismatches in trans-membrane proteins . We designated this tally as the allogenomics mismatch score ( AMS ) . We examined the association between the AMS and post-transplant estimated glomerular filtration rate ( eGFR ) using mixed models , considering transplants from three independent cohorts ( a total of 53 donor-recipient pairs , 106 exomes , and 239 eGFR measurements ) . We found that the AMS has a significant effect on eGFR ( mixed model , effect size across the entire range of the score: -19 . 4 [-37 . 7 , -1 . 1] , P = 0 . 0042 , χ2 = 8 . 1919 , d . f . = 1 ) that is independent of the HLA-A , B , DR matching , donor age , and time post-transplantation . The AMS effect is consistent across the three independent cohorts studied and similar to the strong effect size of donor age . Taken together , these results show that the AMS , a novel tool to quantify amino acid mismatches in trans-membrane proteins in individual donor/recipient pair , is a strong , robust predictor of long-term graft function in kidney transplant recipients .
Survival of patients afflicted with End Stage Renal Disease ( ESRD ) is superior following kidney transplantation compared to dialysis therapy . The short-term outcomes of kidney grafts have steadily improved since the early transplants with refinements in immunosuppressive regimens , use of DNA-based human leukocyte antigen ( HLA ) typing , and better infection prophylaxis [1–3] . Despite these advances , data collected across the USA and Europe show that 40–50% of kidney allografts fail within ten years of transplantation [4] . This observation strongly suggests that as yet uncharacterized factors , including genomic loci , may adversely impact long-term post-transplantation outcomes . The HLA is a cluster of genes on the short arm of chromosome 6 and constitutes the major histocompatibility complex ( MHC ) responsible for self/non-self discrimination in humans . Multiple clinical studies have demonstrated the importance of HLA-matching to improve kidney graft outcome . Therefore , in many countries , including the USA , donor kidney allocation algorithms includes consideration of HLA matching of the kidney recipient and donor . With widespread incorporation of HLA matching in kidney organ allocation decisions , it has become clearer that HLA mismatching represents an important risk factor for kidney allograft failure but fails to fully account for the invariable decline in graft function and failure in a large number of recipients over time . Indeed , only a 15% survival difference exist at 10 years post transplantation between the fully matched kidneys and the kidneys mismatched for both alleles at the HLA-A , B and DR loci [5] . Findings from large cohorts of kidney graft recipients have also been studied to separate the immunological effect mediated by HLA and the non-HLA effects [6] . Overall , prior observations suggest that mismatches at non-HLA loci in the genome could influence long-term graft outcomes . Also , antibodies directed at HLA as well as non-HLA ( e . g . , MHC class I polypeptide-related sequence [MICA] ) have been associated with allograft rejection and reduced graft survival rates . Indeed , it has been reported that the presence of anti-MICA antibodies in the pre-transplant sera is associated with graft failure despite HLA matching of the kidney recipient with the organ donor . Here , we used exome sequencing to determine the sequences of the HLA as well as non-HLA peptides encoded by the donor organ and displayed on its cell surface , as well as bioinformatics analyses to determine donor sequences not present in the recipient . The allogenomics approach integrates the unique features of transplantation , such as the existence of two genomes in a single individual , and the recipient’s immune system mounting an immune response directed at either HLA or non-HLA antigens displayed by the donor kidney . In this report , we show that this new concept helps predict long-term kidney transplant function from the genomic information available prior to transplantation . We found that a statistical model that incorporates time as covariate , HLA , donor age and the AMS ( allogenomics mismatch score , introduced in this study ) , predicts graft function through time better than a model that includes the other factors and covariates , but not the AMS .
The allogenomics concept is the hypothesis that interrogation of the coding regions of the entire genome for both the organ recipient and organ donor DNA can identify the number of incompatible amino-acids ( recognized as non-self by the recipient ) that inversely correlates with long-term function of the kidney allograft . Fig 1A is a schematic illustration of the allogenomics concept . Because human autosomes have two copies of each gene , we consider two possible alleles in each genome of a transplant pair . To this end , we estimate allogenomics score contributions between zero and two , depending on the number of different amino acids that the donor genome encodes for at a given protein position . Fig 1B shows the possible allogenomics score contributions when the amino acids in question are either an alanine , or a phenylalanine or an aspartate amino acid . The allogenomics mismatch score ( AMS ) is a sum of amino acid mismatch contributions . Each contribution represents an allele coding for a protein epitope that the donor organ may express and that the recipient immune system could recognize as non-self ( see Equation 1 and 2 in Fig 1C and Materials and Methods and full description in S1 File ) . We have developed and implemented a computational approach to estimate the AMS from genotypes derived for pairs of recipient and donor genomes . ( See Materials and Methods for a detailed description of this approach and its software implementation , the allogenomics scoring tool , available at http://allogenomics . campagnelab . org . ) Our approach was designed to consider the entire set of protein positions measured by a genotyping assay , or restrict the analysis to a subset of positions P in the genome . In this study , we focused on the subset of genomic sites P that encode for amino acids in trans-membrane proteins . It is possible that some secreted or intra-cellular proteins can contribute to the allogenomics response , but the set of trans-membrane proteins was considered in this study in order to enrich contributions for epitopes likely to be displayed at the surface of donor kidney cells . While proteins expressed in kidney could appear to be a better choice , the technical challenge of defining a list of proteins expressed by kidney alone , and perhaps only transiently in some kidney cell type exposed to the surface of the kidney , argues against relying on a kidney expression filter . Similarly , we did not consider other sets of proteins , and make no claim that the set of transmembrane proteins is an optimal choice . Because the AMS sums contributions from thousands of genomic sites across the genome , it is an example of a burden test , albeit summed across an entire exome . The procedure is akin to averaging and the resulting score is much less sensitive to errors introduced by the genotyping assays or analysis approach than previous association studies which considered genotypes individually . The AMS approach yields a single score per transplant . This eliminates the need to correct for tens of thousands of statistical tests , which are common in classical association studies . The allogenomics approach therefore also decreases the number of samples needed to reach statistical power . In order to test the allogenomics hypothesis , we isolated DNA from kidney graft recipients and their living donors . We assembled three cohorts: Discovery Cohort ( 10 transplant pairs ) where the allogenomics observation was first made ( these patients were a subset of patients enrolled in a multicenter Clinical Trial in Organ Transplantation-04 study of urinary cell mRNA profiling , from whom tissue/cells were collected for future mechanistic studies [7] , 10 transplant pairs ) , and two validation cohorts: one from recipients transplanted at the New York Presbyterian Weill Cornell Medical Center ( Cornell Validation Cohort , 24 pairs ) , and a second validation cohort from recipients transplanted in Paris hospitals ( French Validation Cohort , 19 pairs ) . Table 1 provides demographic and clinical information about the patients included in our study . Exome data were obtained for each cohort . For the Discovery cohort , we used the Illumina TrueSeq exome enrichment kit v3 , covering 62Mb of the human genome . For the two validation cohorts , DNA sequencing was performed using the Agilent Haloplex assay covering 37Mb of the coding sequence of the human genome . Primary sequence data analyses were conducted with GobyWeb [8] ( data and analysis management ) , Last [9] ( alignment to the genome ) and Goby [10] ( genotype calls ) . Table A in S1 File provides statistics of coverage for the exome assays . Kidney graft function is a continuous phenotype and is clinically evaluated by measuring serum creatinine levels or using estimated glomerular filtration rate ( eGFR ) [11] . In this study , kidney graft function was evaluated at several time points for each recipient , with the precise time points varying by cohort . In the discovery cohort , kidney allograft function was measured at 12 , 24 , 36 and 48 months following transplantation using serum creatinine levels and eGFR , calculated using the 2011 MDRD [11] formula . We examined whether the allogenomics mismatch score is associated with post-transplantation allograft function . In Fig 2 , we illustrate the association observed between AMS and creatinine levels or eGFR in the Discovery Cohort . We found positive linear associations between the allogenomics mismatch score and serum creatinine level at 36 months post transplantation ( r2 adj . = 0 . 78 , P = 0 . 002 , n = 10 ) but not at 12 or 24 months following kidney transplantation ( Fig 2A , 2B and 2C ) . We also found a negative linear relationship between the score and eGFR at 36 months post transplantation ( r2 adj . = 0 . 57 , P = 0 . 02 ) but not at 12 or 24 months following kidney transplantation ( Fig 2D , 2E and 2F ) . These findings suggest that in the Discovery cohort the AMS is predictive of long-term graft function . It is also possible that the AMS score would predict short-term graft function , but that more data is needed to detect smaller changes in eGFR at early time points , whereas cumulative effects on graft function become detectable at later time points . Similar observations were made in the two validation cohorts ( see Figures A and B in S1 File ) and discussed in detail in an earlier preprint [12] . In the models presented so far , we have considered the prediction of graft function separately at different time points . An alternative analysis would consider time since transplantation , as well as other established predictors of graft function as covariates in the model . This is particularly useful when studying cohorts where graft function was assessed at several distinct time points ( e . g . , in the French cohort , clinical data describes graft function from 1 to 96 months post transplantation , but few time points have observations for all recipients ) . To implement this alternative analysis , we fit a mixed linear model of the form: eGFR ~ donor age at time of transplant + AMS + T + ( 1|P ) ( Equation 3 ) , where T is the time post-transplantation , measured in months , and ( 1|P ) a random effect which models separate model intercepts for each donor/recipient pairs . To determine the effect of AMS on eGFR , we compared the fit of models that did or did not include the AMS . We found that the effect of AMS is significant ( P = 0 . 0042 , χ2 = 8 . 1919 , d . f . = 1 ) . A similar result was obtained if HLA was also used as a covariate in the model ( i . e . , eGFR ~ donor age at time of transplant + AMS + T + HLA + ( 1|P ) ( Equation 4 ) , comparing model with AMS or without , P = 0 . 038 , χ2 = 4 . 284 , d . f . = 1 ) . In contrast , models that included AMS , but did or did not include the number of ABDR HLA mismatches fit the data equally well ( testing the effect of HLA , P = 0 . 60 , χ2 = 0 . 2737 , d . f . = 1 ) , confirming that the effect of AMS was independent of the number of HLA mismatches . The models of equations 3 and 4 include a random effect for the transplant pair ( 1|P ) term . This term models the differences among pairs , such as level of graft function in the days post-transplantation , as well as correlations between repeated measurements for the same recipient . See Fig C in S1 File for a more direct comparison between AMS and HLA ABDR mismatches . This comparison indicates that there is a moderate correlation between AMS and the number of HLA ABDR mismatches . Taken together , these results indicate that the predictive ability of the AMS effect is mostly independent of the number of ABDR mismatches at the HLA loci . In order to determine if the AMS effect is robust , we fit the model from equation 3 in each cohort independently . The estimates for the AMS effect are shown in Table 2 . Despite a limited amount of data to fit the model in each cohort , the estimates are very similar , strongly suggesting that the AMS effect is robust and can be observed even in small cohorts ( 10 , 19 and 24 transplant pairs ) . In Fig D in S1 File we plot the minor allele frequencies ( MAF ) of the variations that contribute to the AMS in the Discovery and Validation cohorts . We find that many polymorphisms that contribute to the AMS have low MAF , indicating that they are rare in human populations . This point needs to be considered for replication studies . For instance , GWAS genotyping platforms may require adequate imputation to infer polymorphisms with low MAF . Table 3 presents confidence intervals for the parameters of the full model ( equation 4 , including HLA term ) , fit across 53 transplant pairs , as well as the effective range of each of the model predictors . The table shows the expected impact of each predictor on eGFR when this predictor is varied over its range , assuming all other predictors are kept constant . For instance , assume that donor age at time of transplant varies from 20 years old to 80 years old ( range: 60 ) . Across this range , eGFR will decrease by an estimated 28 units as the donor gets older . The AMS effect has an effective range of 1 , 700 and the corresponding eGFR decrease is 19 units . This comparison indicates that the strength of the AMS effect is similar to that of donor age and more than five times larger than the effect of HLA- ABDR mismatches .
While HLA-matching is a necessary requirement for successful hematopoietic cell transplants , full HLA compatibility is not an absolute prerequisite for all types of transplantations as indicated by the thousands of solid organ transplants performed yearly despite lack of full matching between the donor and recipient at the HLA-A , B and DR loci . In view of better patient survival following transplantation compared to dialysis , kidney transplants have become the standard of care for patients with end stage kidney disease and transplants are routinely performed with varying degrees of HLA-class I and II mismatches . Although , graft outcomes improve with better HLA-matching [13] , excellent long-term graft outcomes with stable graft function have been observed in patients with full HLA -ABDR mismatches . The success of these transplants clearly suggests that factors other than HLA compatibility may influence the long-term clinical outcome of kidney allografts . Furthermore , grafts do fail even with the best HLA match [13] , suggesting that antigens other than HLA are targets of alloimmune response . Indeed , several non-HLA antibodies have been identified for renal and cardiac allograft recipients and found detrimental to long-term outcome [14 , 15] . These antibodies were found to target antigens expressed on endothelial and epithelial cells but also on a variety of parenchymal and immune cells and can be measured prior to transplantation . These prior studies support the notion that non-HLA antibodies can influence long-term outcome in transplantation . Recipients of a kidney transplant have two genomes in their body: their germline DNA , and the DNA of the donor . It is clear that a Mendelian genetic transmission mechanism is not at play in transplantation , yet , this assumption has been made in most of the transplantation genomic studies published to date [16 , 17] . While several case-control studies have been conducted with large organ transplant cohorts , the identification of genotype/phenotype associations has been limited to the discoveries of polymorphisms with small effect , that have been reviewed in [18] , and have often not been replicated [19–21] . Rather than focusing on specific genomic sites , the allogenomics concept sums contributions of many mismatches that can impact protein sequence and structure and could engender an immune response in the graft recipient . These allogenomics mismatches , captured in our study , represent the sequences of non-HLA trans-membrane proteins , some of which may help initiate cellular and humoral immunity directed at the allograft . This study used eGFR as a surrogate marker for long-term graft survival . The advantage of focusing on eGFR is that it is measured as part of clinical care on a yearly basis for each recipient , and eGFR has been associated with long-term outcome in multiple studies . Since acute rejection has also been associated with a decrease in long-term graft survival , it may also serve as a surrogate marker for long-term kidney allograft survival . Acute rejection however is a rare event with current immunosuppressive regimens and given the relatively small size of our study cohort , we would not have had sufficient cases to examine the association between acute rejection and the allogenomics score . Another consideration for not using acute rejection is that acute rejection only represents a fraction of the mechanisms that lead to graft loss [22] . The allogenomics concept that we present in this manuscript postulates a mechanism for the development of the immune response in the transplant recipient: immunological and biophysical principles strongly suggest that alleles present in the donor genome , but not in the recipient genome , will have the potential to produce epitopes that the recipient immune system will recognize as non-self . This reasoning explains why the allogenomics score is not equivalent to the genetic measures of allele sharing distance that have been used to perform genetic clustering of individuals [23] . This manuscript also suggests that allogenomic mismatches in proteins expressed at the surface of donor cells could explain why some recipients’ immune systems mount an attack against the donor organ , while other patients tolerate the transplant for many years , when given similar immunosuppressive regimens . If the results of this study are confirmed in additional independent transplant cohorts ( renal transplants , solid or hematopoeitic cell transplants ) , they may prompt the design of prospective clinical trials to evaluate whether allocating organs to recipients with a combination of low allogenomics mismatch scores and different HLA mismatch scores improves long term graft outcome . A positive answer to this question could profoundly impact the current clinical and regulatory framework for assigning organs to ESRD patients . In this study , we introduced the allogenomics concept to quantitatively estimate the histoincompatibility between living donor and recipient outside of the HLA loci . We tested the simplest model derived from this concept to calculate an allogenomics mismatch score ( AMS ) reflecting the possible donor specific epitopes displayed on the cell surface . We demonstrated that the AMS , which can be estimated before transplantation , helps predict post-transplantation kidney graft function more accurately than HLA-mismatches alone . Interestingly , the strength of the correlation increases with the time post transplantation , an intriguing finding observed in both the discovery cohort and the validation cohorts . We chose the simplest model to test the allogenomics concept and did not restrict the score to contributions from the peptides that can fit in the HLA groove despite their computational predictability [24] . It is possible that such restriction would increase the score’s ability to predict renal function post transplantation . However , such a filter assumes that HLA and associated peptides are the only stimuli for the anti-allograft response and does not take into consideration allorecognition involving innate effectors ( NK cells or NKT cells for example , the Killer-cell Immunoglobulin-like Receptor KIR genes , iTCR , the invariant T Cell Receptor , and TLR , Toll Like Receptor , among others ) [25] . The allogenomics concept incorporating amino acid mismatches capable of triggering adaptive as well as innate immunity could be considered an important strength of the approach . Recent evidence indicates that mutations in splice sites , although rare , are responsible for a large proportion of disease risk [26] . The allogenomics approach presented in this manuscript does not incorporate knowledge of how polymorphisms in splice sites affect protein sequences . We anticipate that future developments would consider longer splice forms in the donor as allogenomics . Such an approach could score additional donor protein residues as allogenomics mismatches when the sequence is not present in the predicted proteome of the recipient . We chose to focus this study on living , ABO compatible ( either related or non-related ) donors because kidney transplantation can be planned in advance and because differences in cold ischemia times and other covariates common in deceased donor transplants are negligible when focusing on living donors , especially in small cohorts . The selection criteria for deceased donors include consideration of HLA matching , calculated panel reactive antibody and the age of the recipient . Compared to live donors we expect that the range of the AMS in deceased donors will be comparable to that in our discovery cohort composed primarily of unrelated donors . Since many additional factors can independently influence graft function after transplantation from a deceased donor ( e . g . cold ischemia time ) , potentially much larger cohorts may be required in such settings to achieve sufficient power to adequately control for the covariates relevant to deceased donors and to detect the allogenomics effect . While we have not attempted to optimize the set of sites considered to estimate the allogenomics mismatch score , it is possible that a reduced and more focused subsets of amino acid mismatches could increase the predictive ability of the score . For instance , the AMS could be applied to look for genes with a high allogenomic mismatch burden . Such studies would require larger cohorts and may enable the discovery of loci enriched in allogenomics mismatches responsible for a part of the recipient alloresponse against yet unsuspected donor antigens . Their discovery might foster the development of new immunosuppressive agents targeting the expression of these immuno-dominant epitopes . However , our study also raises a novel mechanistic hypothesis: the total burden of allogenomics mismatches might be more predictive of graft function , than mismatches at specific loci , as was previously widely expected [17] .
The study was reviewed and approved by the Weill Cornell Medical College Institutional Review Board ( protocol #1407015307 “Predicting Long-Term Function of Kidney Allograft by Allogenomics Score” , approved 09/09/2014 ) . The second study involving the French cohort was approved by the Comité de Protection des Personnes ( CPP ) , Ile de France 5 , ( 02/09/2014 ) . Codes were used to ensure donor and recipient anonymity . All subjects gave written informed consent . Living donor ABO compatible kidney transplantations were performed according to common immunological rules for kidney transplantation with a mandatory negative IgG T-cell complement-dependent cytotoxicity cross-match . Briefly , genotypes of donors and recipients were assayed by exome sequencing ( Illumina TruSeq enrichment kit for the Discovery Cohort and Agilent Haloplex kit for the Cornell Validation Cohort and the French Validation Cohort ) . Reads were aligned to the human genome with the Last [9] aligner integrated as a plugin in GobyWeb [8] . Genotype calls were made with Goby [10] and GobyWeb [8] . Prediction of polymorphism impact on the protein sequence were performed with the Variant Effect Predictor [27] . Genes that contain at least one transmembrane segment were identified using Ensembl Biomart [28] . We selected 10 kidney transplant recipients from those who had consented to participate in the Clinical Trials in Organ Transplantation-04 ( CTOT-04 ) , a multicenter observational study of noninvasive diagnosis of renal allograft rejection by urinary cell mRNA profiling . We included only the recipients who had a living donor kidney transplant and along with their donors , had provided informed consent for the use of their stored biological specimens for future research . Pairs were limited to those where enough DNA could be extracted to perform the exome assay for both donor and recipient . Subjects were not selected on the basis of eGFR , whose values were collected after obtaining sequence data . The demographic and clinical information of the Discovery cohort is shown in Table 1 . DNA was extracted from stored peripheral blood using the EZ1 DNA blood kit ( Qiagen ) based on the manufacturer’s recommendation . DNA was enriched for exome regions with the TruSeq exome enrichment kit v3 . Sequencing libraries were constructed using the Illumina TruSeq kit DNA sample preparation kit . Briefly , 1 . 8 μg of genomic DNA was sheared to average fragment size of 200 bp using the Covaris E220 ( Covaris , Woburn , MA , USA ) . Fragments were purified using AmpPureXP beads ( Beckman Coulter , Brae , CA , USA ) to remove small products ( <100 bp ) , yielding 1 μg of material that was end-polished , A-tailed and adapter ligated according to the manufacturer’s protocol . Libraries were subjected to minimal PCR cycling and quantified using the Agilent High Sensitivity DNA assay ( Agilent , Santa Clara , CA , USA ) . Libraries were combined into pools of six for solution phase hybridization using the Illumina ( Illumina , San Diego , CA , USA ) TruSeq Exome Enrichment Kit . Captured libraries were assessed for both quality and yield using the Agilent High Sensitivity DNA assay Library Quantification Kit . Sequencing was performed with six samples per lane using the Illumina HiSeq 2000 sequencer and version 2 of the sequencing-by-synthesis reagents to generate 100 bp single-end reads ( 1×100SE ) . We studied 24 kidney transplant recipients who had a living donor transplant at the NewYork-Presbyterian Weill Cornell Medical Center . This was an independent cohort and none of the recipients had participated in the CTOT-04 trial . Recipients were selected randomly based on the availability of archived paired recipient-donor DNA specimens obtained at the time of transplantation at our Immunogenetics and Transplantation Laboratory . DNA extraction from peripheral blood was done using the EZ1 DNA blood kit ( Qiagen ) based on the manufacturer’s recommendation . We studied 19 kidney transplant recipients who had a living donor transplant at Tenon Hospital . This represented a third independent cohort . Recipients were selected randomly based on the availability of archived paired recipient-donor DNA specimens obtained either at the Laboratoire d'histocompatibilité , Hôpital Saint Louis APHP , Paris or during patient’s follow-up between October 2014 and January 2015 . DNA extraction from peripheral blood was done using the Nucleospin blood L kit ( Macherey-Nagel ) based on the manufacturer’s recommendation . The Cornell and French Validation cohorts were both assayed with the Agilent Haloplex exome sequencing assay . The Haloplex assay enriches 37 Mb of coding sequence in the human genome and was selected for the validation cohort because it provides a strong and consistent exome enrichment efficiency for regions of the genome most likely to contribute to the allogenomics contributions in protein sequences . In contrast , the TrueSeq assay ( used for the Discovery Cohort ) enriches 63Mb of sequence and includes regions in untranslated regions ( 5’ and 3’ UTRs ) , which do not contribute to allogenomics scores and therefore do not need to be sequenced to estimate the score . Libraries were prepared as per the Agilent recommended protocol . Sequencing was performed on an Illumina 2500 sequencer with the 100bp paired-end protocol recommended by Agilent for the Haloplex assay . Libraries were multiplexed 6 per lane to yield approximately 30 million paired end reads per sample . We determined the minor allele frequency of sites used in the calculation of the allogenomics mismatch score using data from the Exome Aggregation Consortium ( ExAC ) . This resource made it possible to estimate MAF for most of the variations that are observed in the subjects included in our discovery and validation cohort . Data was downloaded and analyzed with R and MetaR scripts ( see analysis scripts provided at https://bitbucket . org/campagnelaboratory/allogenomicsanalyses ) . We use the NHLBI Exome Sequencing Project ( ESP ) release ESP6500SI-V2 [30] . The ESP measured genotypes in a population of 6 , 503 individuals across the EA and AA populations using an exome-sequencing assay [30] . Of 12 , 657 sites measured in the validation cohort with an allogenomics contribution strictly larger than zero ( 48 exomes , sites with contributions across 24 clinical pairs of transplants ) , 9 , 765 ( 78% ) have also been reported in ESP ( 6 , 503 exomes ) . Illumina sequence base calling was performed at the Weill Cornell Genomics Core Facility . Sequence data in FASTQ format were converted to the compact-reads format using the Goby framework [14] . Compact-reads were uploaded to the GobyWeb[8] system and aligned to the 1000 genome reference build for the human genome ( corresponding to hg19 , released in February 2009 ) using the Last [9 , 31] aligner ( parallelized in a GobyWeb [8] plugin ) . Single nucleotide polymorphisms ( SNPs ) and small indels genotype were called using GobyWeb with the Goby [32] discover-sequence-variants mode ( parameters: minimum variation support = 3 , minimum number of distinct read indices = 3 ) and annotated using the Variant Effect Predictor [27] ( VEP version 75–75 . 7 ) from Ensembl . The data were downloaded as a Variant Calling format [33] ( VCF ) file from GobyWeb [8] and further processed with the allogenomics scoring tool ( see http://allogenomics . campagnelab . org ) . The allogenomics mismatch score Δ ( r , d ) is estimated for a recipient r and donor d as the sum of score mismatch contributions ( see Fig 1C and supplementary methods in S1 File ) . Analyses were conducted with either JMP Pro version 11 ( SAS Inc . ) or metaR ( http://metaR . campagnelab . org ) . Fig 2 as well as Figures in S1 File were constructed with metaR analysis scripts and edited with Illustrator CS6 to increase some font sizes or adjust the text of some axis labels . The model that includes the time post-transplantation as a covariate was constructed in metaR and JMP . The R implementation of train linear model uses the lm R function . This model was executed using the R language 3 . 1 . 3 ( 2015-03-09 ) packaged in the docker image fac2003/rocker-metar:1 . 4 . 0 ( https://hub . docker . com/r/fac2003/rocker-metar/ ) . Models with random effects were estimated with metaR 1 . 5 . 1 and R ( train mixed model and compare mixed models statements , which use the lme4 R package [34] ) . Comparison of fit for models with random effects was obtained by training each model alternative with REML = FALSE an performing an anova test , as described in [35] . We distribute the code necessary to reproduce most of the analysis presented in this manuscript at https://bitbucket . org/campagnelaboratory/allogenomicsanalyses .
|
The article describes a new concept to help match donor organs to recipients for kidney transplantation . The concept relies on the ability to measure the individual DNA of potential donors and recipients . When the data about genomes ( i . e . , DNA ) of possible donors and recipients are available , the article describes how data can be computationally compared to identify differences in these genomes and quantify the possible future impact of these differences on the functioning of the graft . The concept presented in the article determines a score for each pair of possible donor and recipient . This score is called the allogenomics mismatch score . The study tested the ability of this score to predict graft function ( the ability of the graft to filter blood ) in the recipient several years after transplantation surgery . The study found that , in three small sets of patients tested , the score is a strong predictor of graft function . Prior studies often assumed that only a small number of locations in the genome were most likely to have an impact on graft function , while this study found initial evidence that differences across DNA that code for a large number of proteins can have a combined impact on graft function .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"urinary",
"system",
"procedures",
"medicine",
"and",
"health",
"sciences",
"organ",
"transplantation",
"immunology",
"biomarkers",
"human",
"genomics",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"clinical",
"medicine",
"renal",
"transplantation",
"genome",
"analysis",
"kidneys",
"transplantation",
"immune",
"system",
"proteins",
"proteins",
"creatinine",
"biochemistry",
"anatomy",
"clinical",
"immunology",
"transplantation",
"immunology",
"genetics",
"biology",
"and",
"life",
"sciences",
"renal",
"system",
"genomics",
"computational",
"biology",
"genomic",
"medicine"
] |
2016
|
Exome Sequencing and Prediction of Long-Term Kidney Allograft Function
|
In recent years , increasing associations between microRNAs ( miRNAs ) and human diseases have been identified . Based on accumulating biological data , many computational models for potential miRNA-disease associations inference have been developed , which saves time and expenditure on experimental studies , making great contributions to researching molecular mechanism of human diseases and developing new drugs for disease treatment . In this paper , we proposed a novel computational method named Ensemble of Decision Tree based MiRNA-Disease Association prediction ( EDTMDA ) , which innovatively built a computational framework integrating ensemble learning and dimensionality reduction . For each miRNA-disease pair , the feature vector was extracted by calculating the statistical measures , graph theoretical measures , and matrix factorization results for the miRNA and disease , respectively . Then multiple base learnings were built to yield many decision trees ( DTs ) based on random selection of negative samples and miRNA/disease features . Particularly , Principal Components Analysis was applied to each base learning to reduce feature dimensionality and hence remove the noise or redundancy . Average strategy was adopted for these DTs to get final association scores between miRNAs and diseases . In model performance evaluation , EDTMDA showed AUC of 0 . 9309 in global leave-one-out cross validation ( LOOCV ) and AUC of 0 . 8524 in local LOOCV . Additionally , AUC of 0 . 9192+/-0 . 0009 in 5-fold cross validation proved the model’s reliability and stability . Furthermore , three types of case studies for four human diseases were implemented . As a result , 94% ( Esophageal Neoplasms ) , 86% ( Kidney Neoplasms ) , 96% ( Breast Neoplasms ) and 88% ( Carcinoma Hepatocellular ) of top 50 predicted miRNAs were confirmed by experimental evidences in literature .
MicroRNAs ( miRNAs ) are a kind of endogenous non-coding RNA with the length of about 22 nucleotides , regulating the expression of genes by base paring with target messenger RNA ( mRNA ) [1] . Since the first two miRNAs , lin-14 and let-7 , both showing positive regulation for gene expression , were found [1] , increasing new miRNAs have entered into researchers’ horizons . According to latest miRbase ( Release 22 ) , a miRNA database [2] , 38589 entries representing hairpin precursor miRNAs and 48885 mature miRNA products in 271 species are collected . Accumulative evidences have revealed that miRNAs usually negatively regulate gene expression and they play critical roles in various biological processes such as cell proliferation , differentiation , aging and death [3–7] . In addition , mounting close relations between miRNAs and human diseases were confirmed by abundant experimental reports . For example , the existing study has validated that the expression of mir-140 was reduced in osteoarthritic cartilage [8] . Another example is that down-regulation of mir-145 was related to the increased expression of ERG , over-expression of which was the distinct characteristic of prostate cancer [9] . Besides , deregulation of a set of miRNAs including mir-150 , mir-550 , mir-124a , mir-518b and mir-539 was shown to be associated with transformation of gastritis into extranodal marginal zone lymphoma [10] . It is believed that uncovering more miRNA-disease associations gives an insight into molecular mechanisms of diseases and is favorable to diagnosis , prognosis and treatment of human complex diseases [11 , 12] . However , the existing knowledge of miRNA-disease associations is not enough and known associations were mostly obtained from previous biological experiments that were time-consuming and costly . Therefore , increasing studies were devoted to developing computational models to predict potential miRNA-disease associations [13] . These computational models could infer miRNAs that were more likely to be related to the given disease . Based on the prediction results , biological experiments were preferentially conducted for those miRNAs to improve experimental efficiency and save time as well as expenditure . Base on the known miRNA-disease associations in some well-known biological databases such as HMDD V2 . 0 [14] , dbDEMC [15] and miR2Disease [16] , many computational models were proposed to predict associations between miRNAs and diseases , most of which were under the assumption that functionally similar miRNAs are likely to be associated with phenotypically similar diseases [17–19] . These methods cover two main categories , network algorithm and machine learning . For example , by integrating miRNA functional similarity network , the disease phenotype similarity network and the known disease-miRNA associations network , Jiang et al . [20] proposed initial computational model to uncover potential miRNA-disease associations . For an investigated disease d , each miRNA in the miRNA network was scored by the scoring function based on cumulative hypergeometric distribution . However , the model only considered local neighbor similarity information of each miRNA so that it did not show excellent prediction results . Xuan et al . [21] developed a model of Human Disease-MiRNA association Prediction ( HDMP ) to predict disease-related miRNAs . In this model , miRNA functional similarity was calculated and for miRNAs in the same family or cluster , their similarity scores were given higher weight because they tend to be associated with the same disease . For investigated disease d , relevance score of each miRNA candidate was calculated based on its most weighted k similar neighbors and then ranked to attained potential d-related miRNAs . Nevertheless , HDMP were unable to work for new disease without any known associated miRNAs . In addition , HDMP was also a local network similarity-based model that only considered miRNAs’ partial similarity information , such as neighbor information . In order to make full use of global network similarity information , Chen et al . [22] first adopted global network similarity measures and proposed a method of Random Walk with Restart for MiRNA-Disease Association prediction ( RWRMDA ) in which random walk was implemented on miRNA functional similarity network . Although the model achieved satisfactory prediction performance , it could not deal with new disease without any known associated miRNAs . Another model named MIDP was proposed by Xuan et al . [23] based on random walk on miRNA functional similarity network . Furthermore , MIDPE that was extended from MIDP could predict potential related miRNAs for new disease without any known related miRNAs . Chen et al . [24] proposed the model of Within and Between Score for MiRNA-Disease Association prediction ( WBSMDA ) to predict potential miRNA-disease associations , which specially calculated Gaussian interaction profile kernel similarity for diseases and miRNAs in addition to using the miRNA functional similarity and the disease semantic similarity . In this model , both of the Within-Score and Between-Score were defined from the view of miRNAs and diseases and the final association score for miRNA-disease pair was calculated by combining Within-Score and Between-Score . WBSMDA could also be effectively applied for new diseases and new miRNAs without any known associations . Chen et al . [25] further developed the model of Heterogeneous Graph Inference for MiRNA-Disease Association prediction ( HGIMDA ) in which the heterogeneous graph was constructed with the same inputs as WBSMDA . An iteration process was adopted based on the graph to infer potential miRNA-disease associations . For a further improvement of prediction accuracy , Chen et al . [26] proposed another method named Matrix Decomposition and Heterogeneous Graph Inference for miRNA-disease association prediction ( MDHGI ) , which fully utilized matrix decomposition technique for known miRNA-disease associations before constructing the heterogeneous graph that was same as HGIMDA . In addition , method of Super-Disease and MiRNA for potential MiRNA–Disease Association prediction ( SDMMDA ) was proposed by Chen et al [27] . In order to improve the similarity measures of diseases and miRNAs , the model introduced ‘super-miRNA’ and ‘super-disease’ that were obtained by clustering as many as possible similar miRNAs or diseases . In addition , You et al . [18] proposed the prediction model of Path-Based MiRNA-Disease Association prediction ( PBMDA ) that integrated various biological datasets that was same as MDHGI into the heterogeneous graph . In the graph , the association possibility was calculated by summing all path scores between a miRNA and a disease . Specially , the model penalized long paths by a decay function because these paths were considered to make less contribution to the association score for the miRNA-disease pair . However , the distance-decay function in this model was relatively simple and could be further optimized . Yu et al . [28] proposed the prediction method , MaxFlow , which constructed a miRNAome-phenome network graph where a source node and a sink node were introduced . For the given disease , the maximum information flow from the source over all links to the sink were calculated and flow quantity leaving a miRNA node was used as the association score between the miRNA and the given disease . Furthermore , Chen et al . [29] developed another prediction model of Bipartite Network Projection for MiRNA–Disease Association prediction ( BNPMDA ) . This model first constructed the bias ratings for miRNAs and diseases based on three networks , including the known miRNA–disease association network , the disease similarity network and the miRNA similarity network . Then bipartite network recommendation algorithm was implemented to reveal potential miRNA-disease associations . In fact , many previous computational models were established based on other types of interaction networks , such as protein-protein interaction ( PPI ) network , miRNA-target interaction network and so on . For example , Shi et al . [30] developed prediction model by mapping disease genes and miRNA targets on PPI networks . For a given miRNA and disease , random walk was performed on the network using the disease genes and the miRNA targets as seeds simultaneously to obtain enrichment scores as association scores of the miRNA-disease pairs . Additionally , Mork et al . [31] proposed a model of miRNA-Protein-Disease ( miRPD ) association prediction with proteins as the mediators , which integrated miRNA–protein associations and protein–disease associations to predict novel associations between miRNAs and diseases . However , performance of miRPD was strongly limited by miRNA-target interactions with the high false positive rate . In addition , Pasquier et al . [32] established MiRAI model that represented distributional information of miRNAs and diseases in a high-dimensional vector space and predicted novel miRNA-disease associations in terms of vector similarities . Nowadays , machine learning has been widely applied in biomedical research [33 , 34] , such as drug target prediction [35] , transcription factor binding prediction [36] , functional variant annotation [37] , synergistic drug combination prediction [38] , small molecule-miRNA interaction prediction [39] , association prediction between long non-coding RNAs and diseases [40] , and disease related RNA methylation prediction [41] . Many machine learning-based methods have been proposed to infer potential miRNA-disease associations [13] . Unlike many previous models , the model of Matrix Completion for MiRNA-Disease Association prediction ( MCMDA ) developed by Li et al . [17] was only depended on known miRNA-disease associations where singular value thresholding ( SVT ) algorithm was used to conduct matrix completion procedure and predict new miRNA-disease association . The drawback of MCMDA was that it could not predict miRNAs for new diseases without any associations . Chen et al . [42] proposed a model named Restricted Boltzmann Machine for Multiple types of MiRNA-Disease Association prediction ( RBMMMDA ) to predicted not only novel miRNA-disease associations but also types of association . In RBMMMDA , a two-layer undirected graphical model of Restricted Boltzmann Machine ( RBM ) was constructed and trained to implement prediction . RBMMMDA also could not predict miRNAs for new diseases without any known association information . Xu et al . [43] proposed a method based on a heterogeneous MiRNA-Target Dysregulated Network ( MTDN ) . A classifier named Support Vector Machine ( SVM ) was built to separate positive miRNA-disease associations from negative ones based on features extracted from MTDN . Nevertheless , it was difficult to select accurate negative samples because of unavailable validation for the negative ones . Another model named Regularized Least Squares for MiRNA-Disease Association prediction ( RLSMDA ) that did not need negative samples was developed by Chen et al . [44] . Under the framework of Regularized Least Squares ( RLS ) , cost functions were defined and minimized to yield optimal classifiers from miRNA and disease sides , respectively . Then the weighted average strategy was adopted to combine two optimal classifiers to obtain final prediction results . Furthermore , Chen et al . [27] introduced the model of Ranking-based K-Nearest-Neighbors for MiRNA-Disease Association prediction ( RKNNMDA ) to infer potential associations between miRNAs and diseases . Based on k-nearest-neighbors for miRNAs and diseases , the model calculated Hamming loss to rank these neighbors with SVM and utilized weighted voting to each predicted miRNA-disease association . In addition , Chen et al . [45] proposed another model called Laplacian Regularized Sparse Subspace Learning for MiRNA-Disease Association prediction ( LRSSLMDA ) which achieved prediction scores from miRNA and disease side , respectively . The model’s inputs were miRNA/disease statistical features and graph theoretic features that were extracted from the miRNA/disease similarity . Then objective functions were built in miRNA/disease side with L1-norm constraint and Laplacian regularization terms . Final predictive results were attained by combining optimization results for objective functions . Furthermore , Chen et al . [46] developed the model of Predicting MiRNA–Disease Association based on Inductive Matrix Completion ( IMCMDA ) , which was a matrix completion-based model . MiRNA-disease association matrix was a sparse matrix and missing association values of miRNA-disease pairs could be completed by means of miRNA similarity and disease similarity feature vectors . Considering different limitations of previous models and improvement room for prediction accuracy , we developed the model of Ensemble of Decision Tree based MiRNA-Disease Association prediction ( EDTMDA ) to infer novel miRNA-disease associations . The inputs of the model were features which were extracted from integrated miRNA similarity , disease similarity and known miRNA-disease associations . The model adopted ensemble learning strategy that integrated multiple classifiers ( base learners ) to get final prediction results , which reflected association probability for candidate miRNA-disease pairs . Three cross validation methods , including global leave-one-out cross validation ( LOOCV ) , local LOOCV and 5-fold cross validation ( 5-fold CV ) were implemented to evaluate performance of EDTMDA . As a result , AUC of 0 . 9309 for global LOOCV , 0 . 8524 for local LOOCV and 0 . 9192+/-0 . 0009 for 5-fold CV were obtained . To our knowledge , the AUCs of EDTMDA are higher than almost all previous models . In addition , three types of case studies for important human diseases were further carried out to evaluate the ability to predict miRNAs related with the investigated disease . There were 47 ( Esophageal Neoplasms ) , 43 ( Kidney Neoplasms ) , 48 ( Breast Neoplasms ) and 44 ( Carcinoma Hepatocellular ) of top 50 predictions confirmed by previously published literature . These aforementioned validation experiments proved that EDTMDA is a reliable and excellent model to predict potential miRNA-disease associations .
In our work , known human miRNA-disease associations verified by experimental evidences in literature were obtained from HMDD V2 . 0 which included 5430 associations between 495 miRNAs and 383 diseases [14] . Here , Y ∈ Rnm×nd was used to denote an adjacency matrix , where nm and nd represented the number of miRNAs and diseases , respectively . If miRNA m ( i ) and disease d ( j ) had association according to HMDD V2 . 0 , the element Y ( m ( i ) , d ( j ) ) equaled to 1 , otherwise 0 . MiRNA functional similarity scores could be computed based on the MISIM method proposed by Wang et al . [47] and downloaded from the website: http://www . cuilab . cn/files/images/cuilab/misim . zip . We denoted FS as the score matrix of miRNA functional similarity and the element FS ( m ( i ) , m ( j ) ) represented the functional similarity scores between miRNA m ( i ) and m ( j ) . Disease semantic similarity was computed according to the literature [47] . we download MeSH descriptors from the National Library of Medicine ( http://www . nlm . nih . gov/ ) , from which the relationship of various diseases could be obtained based on disease Directed Acyclic Graph ( DAG ) . For example , a DAG ( D ) = ( D , T ( D ) , E ( D ) ) was used to represent disease D , where T ( D ) was the node set including all parent nodes of disease D and disease D itself , and E ( D ) was defined as the set of edges pointing to child nodes from parent notes . In DAG ( D ) , we defined the semantic value of disease D to DV1 ( D ) as follows: DV1 ( D ) =∑d∈T ( D ) D1D ( d ) ( 1 ) {D1D ( d ) =1ifd=DD1D ( d ) =max{Δ*D1D ( d′ ) |d′∈childrenofd}ifd≠D ( 2 ) where D1D ( d ) represented the contribution of disease d to the semantic value of disease D in DAG ( D ) . As shown in Eq 2 , disease D was the most specific disease in DAG ( D ) and its contribution to the semantic value of itself was set to 1 . Those parent nodes locating farther from node D are more general denominations , having fewer contribution to the semantic value of disease D . To realize that , semantic contribution factor Δ was introduced ( 0< Δ <1 ) and we set Δ = 0 . 5 in this study , referring to the literature [47] . Based on the assumption that two diseases sharing larger parts in their DAGs tend to have higher semantic similarity , the semantic similarity between disease d ( i ) and d ( j ) could be defined as follows: SS1 ( d ( i ) , d ( j ) ) =∑t∈T ( d ( i ) ) ∩T ( d ( j ) ) ( D1d ( i ) ( t ) +D1d ( j ) ( t ) ) DV1 ( d ( i ) ) +DV1 ( d ( j ) ) ( 3 ) In order to obtain more comprehensive and accurate disease semantic similarity assessment , we needed to measure the similarity from different perspectives . Therefore , another model of measuring disease semantic similarity was adopted according to the literature [21] . We considered that the number of disease DAGs that a disease term may appear in are not always the same and for disease terms in the same layer of DAG ( D ) , the disease term appearing in fewer DAGs should be more informative . i . e . , the disease term should have larger semantic contribution to disease D . In this model , semantic contribution of disease d to disease D in DAG ( D ) was defined as follows: D2D ( d ) =−log[thenumberofDAGsincludingdthenumberofdiseases] ( 4 ) Similar to disease semantic similarity model 1 , the semantic value of disease D and semantic similarity between disease d ( i ) and d ( j ) was respectively given as follows: DV2 ( D ) =∑d∈T ( D ) D2D ( d ) ( 5 ) SS2 ( d ( i ) , d ( j ) ) =∑t∈T ( d ( i ) ) ∩T ( d ( j ) ) D2d ( i ) ( t ) +D2d ( j ) ( t ) DV2 ( d ( i ) ) +DV2 ( d ( j ) ) ( 6 ) Two disease semantic similarity models defined semantic contributions of the disease d to disease D in DAG ( D ) in different ways . We defined it based on the theory that those parent nodes locating farther from node D are more general denominations , having fewer contribution to the semantic value of disease D in model 1 , while in model 2 , we defined it by considering that the disease appearing in fewer DAGs should be more special and have larger semantic contribution to disease D . According to the literature [48] , we could calculate Gaussian interaction profile kernel similarity for miRNAs ( diseases ) , which constructed Gaussian kernel with the adjacency matrix Y . Taking miRNA as an example , the Gaussian interaction profile kernel similarity between miRNA m ( i ) and m ( j ) was calculated as follows: GM ( m ( i ) , m ( j ) ) =exp ( −γd‖Y ( m ( i ) , * ) −Y ( m ( j ) , * ) ‖2 ) ( 7 ) Here , Y ( m ( i ) , * ) and Y ( m ( j ) , * ) are the ith and jth row of adjacency matrix Y , respectively , representing interaction information between corresponding miRNA and all diseases . Parameter γd controlled the bandwidth and was set as follows: γd=γ′d/ ( 1nm∑i=1nm‖Y ( mi , * ) ‖2 ) ( 8 ) Analogically , according to the literature [48] , Gaussian interaction profile kernel similarity for diseases could be calculated as follows: GD ( d ( i ) , d ( j ) ) =exp ( −γd‖Y ( * , d ( i ) ) −Y ( * , d ( j ) ) ‖2 ) ( 9 ) γd=γ′d/ ( 1nd∑i=1nd‖Y ( * , d ( i ) ) ‖2 ) ( 10 ) where Y ( * , d ( i ) ) and Y ( * , d ( j ) ) are the ith and jth column of adjacency matrix Y , respectively , meaning interaction information between corresponding disease and all miRNAs . We computed disease semantic similarity based on DAGs of diseases , but we could not get DAGs for all diseases . That is , for the specific disease without DAG , the semantic similarity score between the disease and other diseases could not be computed in both disease semantic similarity models . In order to obtain all disease similarity information , we integrated disease semantic similarity with Gaussian interaction profile kernel similarity according to [24] as follows: SD ( d ( i ) , d ( j ) ) ={SS1 ( d ( i ) , d ( j ) ) +SS2 ( d ( i ) , d ( j ) ) 2d ( i ) andd ( j ) hassemanticsimilarityGD ( d ( i ) , d ( j ) ) otherwise ( 11 ) where the average of two disease semantic similarity models was used as disease semantic similarity . Similarly , integrated miRNA similarity was given according to [24] as follows . EDTMDA was implemented based on integrated miRNA similarity matrix SM , integrated disease similarity matrix SD and known miRNA-disease associations matrix Y . At first , according to literature [49] , three types of miRNA ( disease ) features were extracted based on the above matrixes SM ( SD ) and Y and used to form the feature vectors , represented by FM ( FD ) . Type 1 features covered the statistical measures summarized for each individual miRNA ( disease ) in Y and SM ( SD ) ( including sum , mean , histogram distributions of miRNA/disease similarity scores ) ; type 2 features included graph theoretical measures for network constructed by SM ( SD ) ( including some neighbors’ attributes , betweenness , closeness , eigenvector centrality and Page-Rank scores of miRNA/disease similarity network ) ; type 3 features focused on each miRNA-disease pair in Y based on matrix factorization of Y and graph theory-related statistics for network constructed by Y . Then , ensemble learning strategy was introduced based on random selection of negative samples and features , which included many base learnings and each base learning yield a base classifier , DT . Particularly , PCA was employed to reduce feature dimensionality during each base learning . The final association scores were obtained by computing the average of all prediction results from these DTs ( motivated by the study of Ezzat et al . [50] ) . The base learning contained following steps ( see Fig 1 ) . Firstly , construction of training sample set was operated . Because there were minority positive samples , accounting for about 2 . 9% of all possible samples in HMDD V2 . 0 used by our method , we chose all positive samples and some negative samples which were randomly singled out from the samples without known associations to construct the training set of our model . Particularly , negative samples were guaranteed to have the same number with positive samples . Here , P = { ( m ( i ) , d ( j ) ) |Y ( m ( i ) , d ( j ) ) = 1} and U = { ( m ( i ) , d ( j ) ) |Y ( m ( i ) , d ( j ) ) = 0} represented the set of positive samples and samples with unknown associations , respectively . The set N ( N∈U ) represented negative samples selected from U and |N| = |P| ( |N| and |P| meant the number of elements in N and P , respectively ) . The set of T = P ⋃ N was training set in base learning . In addition , FM∈Rnm×d and FD∈Rnd×d ( d represented the number of extracted miRNA/disease features ) represented feature matrix of miRNAs and diseases in training set T , respectively . We constructed feature subsets of miRNAs and diseases by randomly selecting miRNA/disease features and used parameter r ( 0 < r ≤ 1 ) to control the size of feature subset . That is , ⌊r×d⌋ features were randomly sampled to construct feature subset . FM ( 1 ) ∈Rnm×d1 and FD ( 1 ) ∈Rnd×d1 represented feature subset of miRNAs and diseases , respectively ( where d1 = ⌊r×d⌋ ) . Secondly , feature dimensionality reduction was applied to miRNA/disease feature subset . In our model , ensemble learning strategy was adopted to yield a large number of base learners , which brought much noise or redundant information to degrade prediction performance . To address this issue , PCA , an unsupervised dimensionality reduction algorithm [51] , was employed to reduce miRNA/disease feature dimensionality of feature subset . Here , we saved top 10 miRNA ( disease ) features after dimensionality reduction , keeping almost all feature information . Here , FM ( 2 ) and FD ( 2 ) represented feature matrix of miRNAs and diseases after dimensionality reduction . Thirdly , the DT , a base classifier , was trained with training set . For the sample in training set T , feature principle components of miRNA and disease , i . e . , miRNA feature vector and disease feature vector in FM ( 2 ) and FD ( 2 ) , were spliced as the feature vector of the sample , which was used as input vector of the DT . Our training set could also be denoted with T′ = { ( x1 , y1 ) , ( x2 , y2 ) , ⋯ , ( xn , yn ) } , where xi= ( xi ( 1 ) , xi ( 2 ) ⋯ , xi ( d2 ) ) was the d2-dimensional input vector ( d2 = 20 ) and yi represented the observed value of the ith sample in adjacency matrix Y , and n was the number of samples in training set . For the DT , we constructed the regression tree model with the arithmetic of CART , which was on the basis of squared error minimum criterion [52] . Yielding the regression tree could be described as a progress of building a binary decision tree recursively . If we selected the feature value xi ( j ) to partition feature space R , j and s ( xi ( j ) =s ) were the splitting variable and splitting point , respectively , and two subspaces were defined as follows: R1 ( j , s ) ={x|x ( j ) ≤s}andR2 ( j , s ) ={x|x ( j ) >s} ( 13 ) Then regression tree could be described as: f ( x ) =ckx∈Rk , k=1 , 2 ( 14 ) where ck denoted output value of subspace Rk and its optimal value was calculated by minimizing squared error ∑xi∈Rk ( yi−f ( xi ) ) 2 . The solution was given as follows: c^k=1Nk∑xi∈Rk ( j , s ) yix∈Rm , m=1 , 2 ( 15 ) where Nk was the number of input vectors in subspace Rk . In order to choose the optimal splitting variable and splitting point , variable j and s were traversed to solve the following equation: minj , s[∑xi∈R1 ( j , s ) ( yi−c1 ) 2+∑xi∈R2 ( j , s ) ( yi−c2 ) 2] ( 16 ) The optimal splitting variable j′ and splitting point s′ was obtained . The pair ( j′ , s′ ) was used to partition the feature space according to the formula ( 13 ) and the output was calculated based on the formula ( 14 ) and ( 15 ) . Then new optimal splitting variable and splitting point were sought in subspace R1 and R2 , respectively . Then new output c^k ( k = 1 , 2 , 3 , 4 ) was calculated in 4 subspaces , respectively . This procedure was repeated until the subspace could not be partitioned . At last , the feature space was divided into K subspaces and the final regression tree was described as follows: f ( x ) =ckx∈Rk , k=1 , 2 , ⋯ , K ( 17 ) Based on random selection of negative samples and miRNA/disease features , M base learnings including above three steps were implemented to yield M DTs . The simple average strategy was adopted for these DTs to obtain final prediction scores . Fig 2 shows the pseudocode of EDTMDA . The code and data of EDTMDA is freely available at https://github . com/chi-young1/EDTMDA .
Based on known miRNA-disease associations in HMDD V2 . 0 , we implemented LOOCV and 5-fold CV to evaluate the performance of EDTMDA . Receiver operating characteristic ( ROC ) curves are widely used to evaluate model performance in previous literature of predicting miRNA-disease associations and in order to more conveniently implement performance comparison , we also employed it in our study . Moreover , ROC curves are insensitive to class imbalance , which is suitable for assessing our model’s ability to recover hidden known associations from mass candidates ( unknown associations ) . LOOCV , including global LOOCV and local LOOCV , were implemented to evaluate the performance of EDTMDA . Global LOOCV was used to evaluate model’s global prediction ability for all disease simultaneously , which evaluated recover ability for a hidden miRNA-disease association from all unknown associations . Local LOOCV was used to evaluate model’s local prediction ability for a specific disease , which assessed the recover ability for a hidden miRNA-disease association from unknown associations of the investigated disease . Therefore , there is big difference for these two types of LOOCV . In global LOOCV , each known miRNA-disease association was singled out as test sample in turn and other known associations were treated as training samples for model training . Note that we recalculated Gaussian interaction profile kernel similarity of miRNAs and diseases when a known miRNA-disease association was removed , changing miRNA-disease adjacency matrix . Prediction scores of the test sample and all candidate samples ( That is , those miRNA-disease pairs without association evidences ) could be obtained after implementing EDTMDA . Then the test sample was ranked with all candidate samples based on their scores , and if the rank was higher than the specific threshold , the test sample was successfully predicted . Different from global LOOCV considering all diseases simultaneously , the test sample was only ranked with candidate samples containing the same disease as the test sample . In model performance evaluation , true positive rate ( TPR , sensitivity ) and false positive rate ( FPR , 1-specificity ) are usually calculated based on given threshold . Sensitivity indicates the percentage of the test samples ranked higher than the specific threshold; specificity means the percentage of negative samples ranked below the threshold . When different thresholds were given , we can obtain corresponding TPR and FPR to plot the ROC curve with the TPR as the vertical axis and FPR as the horizontal axis . ROC curve could be used to vividly show predictive performance of the model , and a ROC curve closer to the upper left corner of the figure represents more accurate performance . Furthermore , area under the ROC curve ( AUC ) was calculated to quantitatively evaluate model performance . AUC = 1 represents that the model has perfect prediction performance and AUC = 0 . 5 refers to random performance . We compared the performance of EDTMDA with other classical models in terms of AUC under cross validation . The details of compared models were provided as follows: HGIMDA [25]: The model constructed a heterogeneous graph by integrating multiple biological data , where all paths with the length equal to three were summarized to infer potential miRNA-disease associations ( The parameter used for comparison was α = 0 . 4 ) . MDHGI [26]: The model employed matrix decomposition for miRNA-disease association matrix before implementing the heterogeneous graph inference that was same as HGIMDA ( The parameters used for comparison were α = 0 . 1 , μ = 10−4 , maxμ = 1010 , ρ = 1 . 1 , ε = 10−6 and α = 0 . 4 ) . RLSMDA [44]: The method combined two classifiers trained from the miRNA space and disease space respectively based on the framework of regularized least squares algorithm ( The parameters used for comparison were ηM = 1 , ηD = 1 and ω = 0 . 9 ) . HDMP [21]: The relevance scores of unlabeled miRNAs were computed based on functional similarity of miRNAs’ k nearest neighbors . Besides , the members in the same miRNA family or cluster are assigned higher weight ( The parameters used for comparison were α = 4 , β = 4 and k = 20 ) . WBSMDA [24]: The model defined the Within-Score and Between-Score from the miRNA side and disease side , then combined these score to infer potential miRNA-disease associations . RWRMDA [22]: Random walk was implemented on the miRNA-miRNA functional similarity network ( The parameters used for comparison were r = 0 . 2 and threshold = 10−6 ) . MCMDA [17]: The model utilized the matrix completion algorithm to update the adjacency matrix of known miRNA-disease associations ( The parameters used for comparison were ε = 10-4 and max_iter = 500 ) . MiRAI [32]: The model represented distributional information on miRNAs and diseases in a high-dimensional vector space and defined associations between miRNAs and diseases in terms of their vector similarity ( The parameter used for comparison was r = 400 ) . MaxFlow [28]: A combinatorial prioritization algorithm was designed for miRNA-disease association prediction by modifying the existing maximizing information flow method ( The parameters used for comparison were α = 0 . 1 , β = 0 . 6 , γ = 100 , η = 6 and σ = 10 ) . PBMDA [18]: The model constructed a heterogeneous graph consisting of three interlinked sub-graphs and computed the accumulative contributions from all paths between a miRNA-disease pair as the association score , which specially set decay factor to cut down the contributions of longer paths to miRNA-disease association scores ( The parameters used for comparison were T = 0 . 5 , L = 3 and α = 2 . 26 ) . LRSSLMDA [45]: A common subspace for the miRNA/disease profiles , a L1-norm constraint and Laplacian regularization terms were joint to construct the prediction model ( The parameters used for comparison were γ = 2 , μ = 1 and λ = 1 ) . MIDP [23]: A novel random walk with different transition weight for labeled nodes and unlabeled nodes was implemented on miRNA functional similarity network to predict miRNAs related to the disease with some known related miRNAs and for the new disease without any known related miRNAs , the model extend the walking on a miRNA-disease bilayer network ( The parameters used for comparison were rQ = 0 . 4 and rU = 0 . 1 . ) . Fig 3 showed the performance comparisons between EDTMDA and other several models in the framework of global and local LOOCV . EDTMDA , LRSSLMDA , PBMDA , MDHGI , HGIMDA , MCMDA , MaxFlow , RLSMDA , HDMP and WBSMDA obtained AUC of 0 . 9309 , 0 . 9178 , 0 . 9169 , 0 . 8945 , 0 . 8781 , 0 . 8749 , 0 . 8624 , 0 . 8426 , 0 . 8366 and 0 . 8030 in global LOOCV , respectively; they obtained 0 . 8524 , 0 . 8418 , 0 . 8341 , 0 . 8240 , 0 . 8077 , 0 . 7718 , 0 , 7774 0 . 6953 , 0 . 7702 and 0 . 8031 in global LOOCV , respectively . RWRMDA and MIDP did not have an AUC value in global LOOCV because they could not simultaneously make predictions for all diseases . Additionally , global LOOCV also could not be implemented for MiRAI because the association scores yielded from the model were highly related with the number of known associated miRNAs of a disease . For a disease with more related miRNAs , the association scores for its candidate miRNAs were more likely to be higher . Therefore , it was not objective to simultaneously consider association scores of all diseases in global LOOCV . AUCs of 0 . 7891 for RWRMDA , 0 . 8196 for MIDP and 0 . 6299 for MiRAI were obtained in local LOOCV . Higher AUC values of EDTMDA in LOOCV indicated that our model had more accurate prediction than most previous models . We implemented 5-fold CV to further evaluate the prediction performance of EDTMDA . In 5-fold CV , all positive samples ( That is , those miRNA-disease pairs with known associations ) were randomly divided into five equal-sized groups , four of which , along with same size of selected randomly negative samples , used to training the classifier . The omitted group ( hidden positive samples ) was added to all unknown associations to construct all candidates . Specially , we recalculated the Gaussian interaction profile kernel similarity of miRNAs and diseases when each group of miRNA-disease associations were removed . Then similar to global LOOCV , the association scores of candidates were calculated and then ranked by their scores . The higher the hidden positive samples were ranked , the better the performance was . That is , we removed some known associations and assessed ability to recover these hide associations to evaluate performance of model . This procedure was repeated 100 times because sample division was random in 5-fold CV . As a result , EDTMDA obtained average AUC with standard deviation of 0 . 9192+/-0 . 0009 , surpassing all other methods compared ( See Table 1 ) , which further shows the superior performance of EDTMDA . In our method , multiple base learnings were constructed to generate many base classifiers ( DTs ) base on random selection of negative samples and miRNA/disease features , which also brought some noise or redundancy to influence final prediction results . To address this issue , we used PCA to implement dimensionality reduction for miRNA/disease feature subset . To evaluate the effect of dimensionality reduction to our model , we assessed performance of the method after removing dimensionality reduction step in each base learning . That is , we spliced miRNA and disease features of feature subset as the input of base classifiers . The AUC comparison results between EDTMDA with dimensionality reduction and EDTMDA without dimensionality reduction were shown in Table 2 , which indicated that dimensionality reduction in base learning contributed to improve prediction performance of the model . We conducted comparison of prediction performance between EDTMDA and RF which is also an ensemble learning method with DT as base classifier . Extracted miRNA features and disease features were spliced as the input vector of RF and RF was implemented using RandomForestRegressor that is an algorithm package of RF in Python , where default parameter values were used other than n_estimators ( It was set as 50 , meaning that the number of trees in RF is same as in EDTMDA ) . As shown in Table 3 , EDTMDA is notably outperformed RF under three cross validations . There are two main differences between EDTMDA and RF . First , EDTMDA randomly selected a different negative sample set for each base classifier while RF performed bagging on the same negative set . That is , EDTMDA used more negative samples for model training than RF . Second , EDTMDA included all positive samples in training set for each base classifier , but RF performed bagging on the positive samples so that each DT in RF used only a subset of all positive samples . We concluded that prediction performance of the model was sensitive to positive samples and the best strategy was to include all positive samples for each base classifier in ensemble learning . Moreover , EDTMDA incorporated more data for model training , obtaining better prediction performance than RF . To further access the prediction ability of EDTMDA , three types of case studies were carried out . For the sake of brevity , we selected several important human diseases to analyze in detail . The first type of case study was concerned with Esophageal Neoplasms and Kidney Neoplasms , and known miRNA-disease associations in HMDD V2 . 0 were used as training samples . All candidate miRNAs that were unassociated with the investigated disease in HMDD V2 . 0 were ranked according to their predicted association scores . Top 50 of candidate miRNAs were validated in two other miRNA-disease association databases , dbDEMC [15] and miR2Disease [16] . Esophageal Neoplasms is a serious malignancy with high mortality rate , ranking sixth among all cancer in mortality [53] . Squamous cell carcinoma ( SCC ) is the most common type of Esophageal Neoplasms and the black with SCC was three times higher than the white [54] . There will be 17190 new cases in Esophageal Neoplasms and 15850 people dying of the Esophageal Neoplasms in 2018 according to the study [55] . Many previous studies have confirmed the associations between the Esophageal Neoplasms and various miRNAs . For example , the higher expression of miRNA-506 was found in squamous cell carcinoma ( ESCC ) patients than in heathy people [56] . Moreover , according to the study [57] , the expression of miRNA-382-5p notably increased and miRNA-133a-3p notably decreased in esophageal adenocarcinoma ( EAC ) . In case study of Esophageal Neoplasms , 10 out of top 10 and 47 out of top 50 predicted miRNAs related to Esophageal Neoplasms were confirmed by dbDEMC or miR2Disease ( See Table 4 ) . Kidney Neoplasms , also known as Renal cell carcinoma ( RCC ) , accounts for 2–3% of all the adult cancers [58] . It has been estimated that 65340 Americans will be diagnosed with Kidney Neoplasms and 14970 will die of the disease in 2018 [55] . Some studies have confirmed that dysregulation of miRNAs is closely related to Kidney Neoplasms . For example , Arai et al . [59] found that low expression of mir-10a-5p had association with overall survival in Kidney Neoplasms patients because downregulation of mir-10a-5p inhibited cancer cell migration and invasion . Another study showed that mir-21 played an important role in Kidney Neoplasms progression and could resist chemotherapeutic drugs used for treatment of Kidney Neoplasms [60] . As a result of case study for Kidney Neoplasms , 9 out of the top 10 and 43 out of the top 50 miRNAs were validated to have associations with Kidney Neoplasms by dbDEMC and miR2Disease ( See Table 5 ) . We exhibited complete prediction results inferring potential disease-associated miRNAs that were ranked based on their predicted association scores , which we expect to be beneficial for experimental studies in the future ( See S1 Table ) . The second type of case study for Breast Neoplasms was implemented to prove the applicability of EDTMDA to new diseases without known related miRNAs . We removed all known Breast Neoplasms-miRNA associations in HMDD V2 . 0 so Breast Neoplasms could be regarded as new disease . After implementing EDTMDA to predict and rank potential Breast Neoplasms-related miRNAs based on other known disease-miRNA associations , we confirmed that 10 out of top 10 and 48 out of top 50 predicted Breast Neoplasms-related miRNAs were validated by HMDD V2 . 0 , dbDEMC and miR2Disease ( See Table 6 ) . Hsa-mir-210 , ranking first in our prediction result list , had the greatest possibility associating with Breast Neoplasms . The study of Zehentmayr et al . [61] has revealed the association that hsa-mir-210 was overexpressed in contralateral unaffected breasts ( CUB ) of patients with breast cancer . This case study showed that our model was also reliable when applied to predict miRNAs related with new diseases . Finally , to test robustness of our model , we carried out the third case study for Carcinoma Hepatocellular based on known associations in HMDD V1 . 0 including 1395 associations between 271 miRNAs and 137 diseases . In this case study , we ranked candidate miRNAs for Carcinoma Hepatocellular and validated top 50 predictions with experimental evidences . As has been defined , a candidate miRNA was a miRNA unassociated with the Carcinoma Hepatocellular according to HMDD v1 . 0 , which guaranteed that validation of the predictions was completely independent of training database HMDD V1 . 0 . As a result , 10 out of top 10 and 44 out of top 50 potential miRNAs associated with Carcinoma Hepatocellular were validated by HMDD V2 . 0 , dbDEMC and miR2Disease ( See Table 7 ) . For example , hsa-mir-146b ( 1st in the prediction list ) was down-regulated in Carcinoma Hepatocellular and could inhibit tumor growth and metastasis of Carcinoma Hepatocellular [62] . Aforementioned results indicate that EDTMDA has good robustness , showing satisfactory performance in different dataset . We randomly shuffled ‘1’ and ‘0’ elements and kept their respective numbers unchanged in adjacency matrix , which was used to test whether our model suffered from overfitting . The AUC of three cross validations including global LOOCV , local LOOCV and 5-fold CV were 0 . 4939 , 0 . 4413 and 0 . 5005+/-0 . 0029 respectively , which indicated that EDTMDA effectively avoided overfitting . Furthermore , label randomization test was implemented in three case studies by randomly shuffling ‘1’ and ‘0’ elements and keeping their respective numbers unchanged in adjacency matrix . The results were shown in Table 8 , compared with the results under true labels . From the comparison results , we could draw the conclusion that EDTMDA is an effective tool to unveil more potential miRNAs related to diseases . In our model , we randomly selected some miRNA-disease pairs without known associations as negative samples . Moreover , considering that different diseases with different numbers of associated miRNAs , we designed a new way to select negative samples , which reflected the contribution of each disease to the positive sample set . For the new way , negative samples were sampled randomly for each disease to have the same size as the positive samples of the disease . That is , more negative samples were sampled for the disease with more known associated miRNAs . This new way to select negative samples was named local random and the previous way to select negative samples from all the negative was named global random . For the model using local random to select negative samples , we implemented model evaluation under three cross validations ( global LOOCV , local LOOCV and 5-fold CV ) , and the AUCs were 0 . 8224 , 0 . 7871 and 0 . 8180+/-0 . 0019 respectively , which was significantly inferior to AUCs of 0 . 9309 , 0 . 8524 and 0 . 9192+/-0 . 0009 in our model using global random to select negative samples . For the local random to select negative samples , the poor performance of model could be that more false negative samples ( miRNA-disease pairs with potential associations ) were selected . It is apparently observed that miRNAs prefer to relate to some specific diseases in our dataset and we think that there should be more potential miRNA-disease associations for these specific diseases . But in local random to select negative samples , more selected negative samples were derived from the negative of those specific diseases with more related miRNAs , i . e . , more false negative samples were selected . In global random to select negative samples , we avoided selecting more false negative samples for model training and obtained better model performance .
Increasing researchers are devoted to developing computational methods to infer potential miRNA-disease associations as these methods can be valuable complements to experiments . In this study , we proposed a computational method called EDTMDA under the framework of ensemble learning and dimensionality reduction . The Gaussian interaction profile kernel similarity scores for miRNAs and diseases were first calculated from known miRNA-disease associations . Then integrated miRNA ( disease ) similarity could be obtained via integrating miRNA functional similarity ( disease semantic similarity ) and Gaussian interaction profile kernel similarity of miRNAs ( diseases ) . In addition , the feature vectors for the miRNA-disease pair was constructed by conducting feature extraction on integrated similarity and known miRNA-disease associations . Multiple base learnings were built based on random selection of negative samples and miRNA/disease features so that many decision trees ( DTs , base classifiers ) were attained . Particularly , in order to remove the noise or redundancy , PCA was utilized to reduce feature dimensionality during each base learning . Final prediction results were given by adopting simple average strategy for these DTs . The success of this model is mainly due to the following points . First , comprehensive statistical features , graph theoretic features and matrix factorization results were extracted from similarity information and known associations so that informative input features for the model could be obtained . Furthermore , because feature profiles made the most of similarity and known associations , EDTMDA could work for new diseases without known association information . Second , ensemble learning was designed to integrate multiple basic classifiers for more accurate prediction . In addition , feature dimensionality reduction with PCA could remove noise or redundancy to further improve prediction performance . Third , for the base classifier , the regression tree model with the arithmetic of Classification and Regression Tree ( CART ) was selected in our model , which was the binary tree with simple structure and could avoid the data fragmentation existing in multi-branching tree . However , there were several limitations in our prediction model . To begin with , known miRNA-disease associations were inadequate ( with only 2 . 86% of 189 , 585 miRNA-disease pairs being labeled ) and increasing associations confirmed by experiments in the future would further improve model performance . Additionally , similarity calculation of miRNA and disease in this study may not be perfect and we expect more biological information would be incorporated into similarity measurement . Moreover , EDTMDA might cause bias to miRNAs which have more associated disease records . Finally , negative samples ( miRNA-disease pairs without associations ) were needed in our model . We randomly sampled some pairs without known associations as negative samples for model training . In order to reduce bias and improve prediction performance , multiple base classifiers were trained and integrated . Moreover , dimensionality reduction was employed for each base classifier to reduce noise and redundant information , which further improve performance of model . Actually , it is still difficult to obtain true negative samples ( That is , miRNA-disease pairs show no evidence of association ) , because these true negative samples are scarcely reported in literature . We will make efforts to develop the new approach to identify reliable negative samples in the future .
|
MiRNAs are known as gene regulators and play critical roles in various biological processes . Many associations between miRNAs and human diseases have been identified , which promotes the understanding towards the molecular mechanisms of diseases and contributes to prevention and treatment of diseases . Computational methods of predicting potential miRNA-disease associations make the discovery more efficient and experiments more productive . We developed EDTMDA by constructing a computational framework integrating ensemble learning and dimensionality reduction . We performed global LOOCV , local LOOCV and 5-fold cross validation to evaluate performance of EDTMDA , which outperformed many classic methods . In addition , we carried out three types of case studies on important diseases , which were used to evaluate performance of model based on known associations in HMDD V2 . 0 , for new diseases without known associations and based on known associations in HMDD V1 . 0 . As a result , most predicted miRNAs in top 50 predictions were confirmed by experimental evidences in literature . So , we believe that EDTMDA can make reliable predictions and guide experiments to uncover more miRNA-disease associations .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"linguistics",
"engineering",
"and",
"technology",
"natural",
"antisense",
"transcripts",
"gene",
"regulation",
"carcinomas",
"cancers",
"and",
"neoplasms",
"social",
"sciences",
"gastrointestinal",
"tumors",
"liver",
"diseases",
"decision",
"tree",
"learning",
"oncology",
"decision",
"analysis",
"micrornas",
"management",
"engineering",
"artificial",
"intelligence",
"gastroenterology",
"and",
"hepatology",
"kidneys",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"decision",
"trees",
"gene",
"expression",
"disease",
"vectors",
"hepatocellular",
"carcinoma",
"biochemistry",
"rna",
"anatomy",
"nucleic",
"acids",
"semantics",
"genetics",
"biology",
"and",
"life",
"sciences",
"renal",
"system",
"species",
"interactions",
"non-coding",
"rna",
"machine",
"learning",
"neoplasms"
] |
2019
|
Ensemble of decision tree reveals potential miRNA-disease associations
|
Hantaviruses cause hemorrhagic fever with renal syndrome ( HFRS ) and hantavirus cardio-pulmonary syndrome ( HCPS; also called hantavirus pulmonary syndrome ( HPS ) ) , both human diseases with high case-fatality rates . Endothelial cells are the main targets for hantaviruses . An intriguing observation in patients with HFRS and HCPS is that on one hand the virus infection leads to strong activation of CD8 T cells and NK cells , on the other hand no obvious destruction of infected endothelial cells is observed . Here , we provide an explanation for this dichotomy by showing that hantavirus-infected endothelial cells are protected from cytotoxic lymphocyte-mediated induction of apoptosis . When dissecting potential mechanisms behind this phenomenon , we discovered that the hantavirus nucleocapsid protein inhibits the enzymatic activity of both granzyme B and caspase 3 . This provides a tentative explanation for the hantavirus-mediated block of cytotoxic granule-mediated apoptosis-induction , and hence the protection of infected cells from cytotoxic lymphocytes . These findings may explain why infected endothelial cells in hantavirus-infected patients are not destroyed by the strong cytotoxic lymphocyte response .
Hantaviruses are emerging zoonotic viruses that cause two severe diseases: hemorrhagic fever with renal syndrome ( HFRS ) in Eurasia and hantavirus cardio-pulmonary syndrome ( HCPS; also called hantavirus pulmonary syndrome ( HPS ) ) in the Americas , with case-fatality rates of up to 10% for HFRS and up to 40% for HCPS [1] , [2] . Several different hantaviruses cause HFRS and HCPS . Among them , Hantaan virus ( HTNV ) and Andes virus ( ANDV ) are the most common HFRS- and HCPS-causing hantaviruses , respectively [1] , [2] . Endothelial cells are the main targets for hantaviruses and increased vascular permeability is , as in other hemorrhagic fevers [3] , a hallmark of HFRS and HCPS . The underlying mechanisms of the increased vascular permeability observed in HFRS and HCPS are , however , not completely understood . For example , it is unclear whether hantaviruses themselves or , alternatively , the related immune responses , are responsible for causing pathology [1] , [4]–[6] . Strong CD8 T cell responses are observed in hantavirus-infected patients [7] , [8] . Recent data have also demonstrated that HFRS-patients exhibit a rapid expansion of activated natural killer ( NK ) cells that in many patients persist at elevated numbers for a prolonged period of time [9] . However , autopsies performed on deceased patients have not revealed any obvious damage of hantavirus-infected endothelial cells [4] , [10]–[12] . This contradiction suggests that hantaviruses might possess mechanisms to prevent cytotoxic lymphocytes from killing infected endothelial cells . This reasoning led us to address how hantavirus-infected endothelial cells are affected by cytotoxic lymphocytes . The cytotoxic granule-dependent pathway , involving granzyme B-mediated activation of caspase 3 , is the main pathway used by cytotoxic lymphocytes , including CD8 T cells and NK cells , to induce apoptosis in virus-infected cells [13] , [14] . Here , we show that hantavirus-infected endothelial cells survive exposure to NK cells , cytotoxic lymphocytes preloaded with large amounts of perforin and granzymes [15] . In an analysis of possible mechanisms behind this phenomenon , the hantavirus nucleocapsid protein was found to inhibit the function of granzyme B and caspase 3 .
To study the responses of cytotoxic lymphocytes upon recognition of hantavirus-infected endothelial cells , HTNV-infected and uninfected primary endothelial cells were exposed to peripheral blood-derived short-term IL-2 activated NK cells . HLA class I molecules , ligands for NK cell-inhibitory receptors [16] , were blocked on the infected and uninfected endothelial cells to allow for maximal NK cell-responses . First , NK cell-degranulation towards infected and uninfected endothelial cells was assessed by measurement of surface expression of CD107a , a surrogate marker for lymphocyte degranulation [17] . Similar levels of degranulation were observed for NK cells co-incubated with infected and uninfected endothelial cells ( Figures 1A–B ) , suggesting that the target cells were exposed to similar levels of cytolytic granule content . This prompted us to study the effects of NK cell interaction with the endothelial cells directly . Strikingly , while uninfected endothelial cells were killed , infected endothelial cells survived ( Figure 1C ) . This indicated to us that hantavirus-infected endothelial cells might be protected from NK cell-mediated induction of apoptosis . In support of this notion , exposure of infected endothelial cells to NK cells resulted in virtually no signs of apoptosis assessed by TUNEL-staining , while a marked induction of apoptosis was apparent in uninfected endothelial cells under the same conditions ( Figures 1D–E ) . Although surprising , the present findings corroborated earlier findings showing that hantavirus-infected endothelial cells in patient material , despite strong cytotoxic lymphocyte responses , are generally not damaged [4] , [7]–[12] . The finding that hantaviruses inhibited cytotoxic lymphocyte-mediated cell death downstream of degranulation suggested the possibility that these viruses interfere with induction of apoptosis in infected cells . This led us to first test if hantaviruses could inhibit apoptosis in general , i . e . , not only specifically induced by cytotoxic lymphocytes . Staurosporine is a broad kinase-inhibitor that activates several different apoptosis-inducing pathways [18] , [19] . Endothelial cells infected with ANDV or HTNV were thus exposed to staurosporine and then monitored for apoptosis . Compared to uninfected cells , both ANDV-infected and HTNV-infected cells showed clearly lower levels of apoptosis ( Figure 2A ) . To test if hantavirus-mediated inhibition of staurosporine-induced apoptosis could be observed also in cell types other than endothelial cells , we examined the effects of staurosporine-induced apoptosis in the human lung epithelial cell line A549 . Likewise , both ANDV- and HTNV-infected epithelial cells showed clearly lower levels of apoptosis compared to uninfected cells ( Figure 2B ) . This suggests that hantavirus-mediated inhibition of apoptosis is not limited to cytotoxic lymphocyte-mediated apoptosis , nor is it restricted to endothelial cells . Induction of apoptosis by the cytotoxic granule-dependent pathway and by staurosporine eventually converges into cleavage of pro-caspase 3 into active caspase 3 [14] , [19] , which in turn cleaves cellular proteins leading to apoptosis . The cleaved , activated , form of caspase 3 could be observed early after treatment with staurosporine in infected cells ( Figure 2C ) , showing that hantavirus infection does not completely block initial steps in the process leading to apoptosis induced by staurosporine . However , in line with the observed inhibition of apoptosis in infected cells ( Figure 2A–B ) , significantly lower levels of the cleaved , activated , form of caspase 3 were observed in infected , compared to uninfected , cells ( Figure 2C ) , suggesting that hantavirus can at least partially inhibit activation of caspase 3 . As expected , this was mirrored by significantly lower levels of caspase 3-activity ( Figure 2D ) , and less caspase 3-cleaved PARP ( Figure 2E ) in infected , compared to uninfected , staurosporine-treated cells . Certain viruses encode proteins that antagonize caspase enzyme-function either by interacting with their active site ( direct inhibition ) , leading to cleavage of the viral protein and inhibition of the caspase , or by acting as competitive inhibitors of proteins needed for caspase activation ( indirect inhibition ) [20] . Thus , we hypothesized that inhibition of caspase 3 might be one mechanism used by hantaviruses to prevent infected cells from undergoing apoptosis . Although the canonical caspase 3 cleavage-site DEVD is not present in any hantavirus protein , initial in silico analysis revealed that the nucleocapsid protein , found at high levels in the cytoplasm of infected cells [4] , contains putative cleavage sites for caspase 3 ( e . g . , see Figure S1 for in silico data of ANDV nucleocapsid protein ) . Indeed , in support of this , co-incubation of ANDV nucleocapsid proteins with recombinant active caspase 3 resulted in formation of a novel nucleocapsid protein-fragment with a molecular mass of approximately 35 kDa ( Figure 3A ) . To test if the nucleocapsid protein was cleaved in a similar manner in hantavirus-infected cells , these cells were treated with staurosporine followed by analyses of the nucleocapsid protein cleavage-pattern . Interestingly , a fragment of the nucleocapsid protein with a molecular mass equivalent to that observed in Figure 3A was detected also in staurosporine-treated infected cells ( Figure 3B ) . Importantly , this specific fragment was not observed in staurosporine-treated cells in the presence of the caspase-inhibitor Z-VAD ( Figure 3B ) , suggesting that the nucleocapsid protein is a natural caspase 3-target . This finding prompted us to seek for the ANDV nucleocapsid caspase 3-cleavage site . N-terminal sequencing of the caspase 3-cleaved ANDV nucleocapsid protein fragments revealed that the cleavage occurred after Asp285 , one of the in silico-predicted caspase 3-cleavage sites ( Figure S1 ) , and in line with the observed approximately 35 kDa product observed in Figure 3A and 3B . Furthermore , recombinant caspase 3 was also able to cleave the nucleocapsid protein in lysates from cells transfected with a plasmid expressing the wild-type ANDV nucleocapsid protein , but not in lysates from cells transfected with a plasmid expressing an Asp285 to Ala285-mutated variant of the ANDV nucleocapsid protein ( data not shown ) , verifying the presence of the identified caspase 3-cleavage site . We next analyzed if the nucleocapsid protein could inhibit caspase 3 . Active recombinant caspase 3 was pre-incubated with recombinant ANDV nucleocapsid protein , or with a control protein , and caspase 3-activity was then measured . Pre-incubation of caspase 3 with ANDV nucleocapsid protein , as compared to the control protein , did indeed result in a measurable inhibition of caspase 3-activity ( Figure 3C–D ) . Finally , to confirm that the identified caspase 3-cleavage site in the nucleocapsid protein was responsible for inhibiting caspase 3-activity , we transfected A549 cells either with wild-type ANDV nucleocapsid protein or with the Asp285 to Ala285-mutated variant , and subsequently exposed the cells to staurosporine . Transfection of cells with wild-type ANDV nucleocapsid protein rendered cells significantly less susceptible to apoptosis induction ( Figure 3E ) and displayed reduced caspase 3 activity ( Figure 3F ) . The apoptosis-resistance as well as the caspase 3-activity was reverted when the caspase 3-cleavage site was mutated ( Figure 3E–F ) . Taken together , the results show that hantavirus nucleocapsid protein significantly inhibits apoptosis and as such the protein represents a new viral caspase 3-inhibitor , similar to baculovirus p35 and p49 and poxvirus cytokine response modifier A ( CrmA ) [20] . Cytotoxic lymphocytes use the cytotoxic granule-pathway to induce apoptosis in virus-infected cells . During this process , granzyme B activates caspase 3 . However , granzyme B might also promote cell death independent of caspase 3 [21] , suggesting that inhibition of caspase 3 alone may not fully account for the failure of NK cells to kill hantavirus-infected cells . Two viral proteins , the 100K assembly protein of human adenovirus type 5 ( Ad5-100K ) and poxvirus CrmA , were previously shown to inhibit granzyme B [20] . Interestingly , CrmA inhibited both caspase 3 and granzyme B [20] , [22]–[24] . These reports led us to investigate if the hantavirus nucleocapsid protein could represent a previously unknown virus-encoded protein capable of also inhibiting granzyme B . Incubation of ANDV nucleocapsid protein with recombinant active granzyme B resulted in cleavage of the nucleocapsid protein into at least three fragments ( Figure 4A ) , showing that these two proteins can interact . As evident by the enzyme-specific cleavage patterns ( Figure S2 ) , caspase 3 and granzyme B targeted different sites in the nucleocapsid protein , showing that the nucleocapsid protein likely contains multiple enzyme-specific cleavage sites . Interestingly , pre-incubation with recombinant ANDV nucleocapsid protein resulted , strikingly , in an almost complete inhibition of granzyme B activity ( Figure 4B and C ) . Although the inhibitory effect somewhat decreased over time , almost 90% inhibition remained even after 27 hours of pre-incubation ( Figure 4D ) , suggesting that the nucleocapsid protein is a potent viral granzyme B-inhibitor . As mentioned , NK cells normally kill infected cells through cytotoxic granule-mediated induction of apoptosis , involving granzyme B-mediated activation of caspase 3 [13] , [14] , [20] . To test if hantavirus inhibited NK cell-activation of caspase 3 , we exposed uninfected and HTNV-infected endothelial cells to IL-2 activated NK cells for two hours , and then analyzed caspase 3-activity in the endothelial cells . Indeed , significantly fewer infected , than uninfected , endothelial cells were positive for active caspase 3 ( Fig . 5A–B ) , showing that NK cells fail to activate caspase 3 in hantavirus-infected cells .
Cytotoxic lymphocytes play important roles in host responses against virus-infected cells , mainly via their unique ability to kill infected cells through cytotoxic granule-mediated induction of apoptosis [13] , [14] , [20] . Consequently , some viruses have evolved evasion strategies specifically targeting this pathway of apoptosis-induction [14] , [20] , [25]–[28] . However , only two viruses have previously been reported to encode proteins that specifically inhibit granzyme B-activity . The human adenovirus type 5 encoded Ad5-100K specifically inhibits granzyme B , whereas the poxvirus CrmA inhibits granzyme B , caspase 3 and other caspases [20] , [22]–[24] , [29] . Adenovirus and poxvirus are DNA-viruses expressing 20 to 200 viral proteins; in contrast hantaviruses are small RNA viruses encoding only four structural proteins [5] . In this regard , it is striking that hantaviruses harbor such a specific viral evasion strategy . It may suggest that hantavirus-mediated inhibition of granzyme B , and hence of cytotoxic lymphocyte-mediated killing of infected cells , is a crucial component in the life cycle of hantavirus infection . In addition to activating apoptosis via caspase 3 , granzyme B can induce cell death in caspase 3-independent manners , e . g . , via cleavage of the Rho-associated coiled coil-containing protein kinase II [30] , and might also possess direct antiviral functions [28] , [31]–[34] . Therefore , strong suppression of granzyme B-activity is likely needed for efficient inhibition of its functions . Some of the granzyme B that reaches the infected target cell are found in the nucleus [14] . This translocation of granzyme B from the cytoplasm to the nucleus is believed to involve importin α [32] , [35] . The nucleocapsid protein of some hantaviruses , like HTNV , can bind importin α and prevent its shuttling capacity [36] . It has been speculated that other hantaviruses , like ANDV , might also have this potential [37] . Hence , the nucleocapsid protein might both directly inhibit the function of granzyme B but also indirectly by inhibiting its functions in the nucleus [14] . CD8 T cells have been shown to be able to kill target cells that express hantavirus nucleocapsid protein [38]–[40] . In these studies the nucleocapsid protein was expressed in the target cells using viral vectors . However , to the best of our knowledge , it has never been shown that CD8 T cells can efficiently kill hantavirus-infected target cells . Like NK cells , CD8 T cells primarily rely on the cytotoxic granule-dependent pathway to kill virus-infected cells [13] , [14] , [41] . Here , we observed that NK cells are largely incapable of killing hantavirus-infected endothelial cells , strongly suggesting that CD8 T cells will not be able to efficiently kill hantavirus-infected endothelial cells either . This raises the question why cells expressing the nucleocapsid protein were not protected against CD8 T cells in past studies [38]–[40] . It can be speculated ( i ) that the amount , and ( ii ) that the localization of the nucleocapsid protein might differ between transfected and infected cells , and furthermore ( iii ) that other hantavirus and/or cellular factors might be needed for efficient inhibition of cytotoxic lymphocyte-mediated apoptosis by providing hereto unknown anti-apoptotic mechanism ( s ) that act in parallel to , or in combination with , the nucleocapsid protein . The findings that hantavirus inhibits staurosporine-induced apoptosis and that the nucleocapsid protein inhibited caspase 3-activity and apoptosis suggest that hantavirus can interfere with induction of apoptosis in general . However , the observed caspase 3-inhibition may not by itself explain the strong anti-apoptotic effect observed in staurosporine-exposed hantavirus-infected cells . It is possible that hantavirus might affect induction of apoptosis also upstream of caspase 3 . Possible additional apoptosis-inducing pathway ( s ) that hantaviruses may inhibit remains to be identified . This is also true for the observed protection against NK cells; while inhibition of granzyme B and caspase 3 is likely to explain a significant part of the hantavirus-mediated protection of infected cells against NK cells , it is possible that also other , hereto unknown , mechanisms are involved . Furthermore , an interesting question is if different hantaviruses use the same mechanisms to inhibit apoptosis . We observed strong inhibition against staurosporine-induced apoptosis and against cytotoxic lymphocyte-killing both in ANDV and HTNV-infected cells , showing that they share the capability to inhibit apoptosis . However , the caspase 3-cleavage site , DLID285 , detected in ANDV nucleocapsid protein is not conserved among hantaviruses , and while the HTNV nucleocapsid protein , as well as nucleocapsid proteins encoded by other hantaviruses , also contain several in silico predicted caspase 3 and granzyme B-cleavage sites ( our unpublished observation ) , it remains to be shown if they represent functional cleavage sites and if they are involved in inhibition of caspase 3 and/or granzyme B . We have previously observed increased levels of caspase-cleaved cytokeratin 18 , a specific marker for epithelial cell apoptosis , during the acute phase of HFRS [42] . While endothelial cells are the main targets for hantavirus , epithelial cells are also infected . The present study demonstrates that hantavirus efficiently inhibits apoptosis also in infected epithelial cells . Interestingly , in this context , it has previously been reported that hantavirus induces apoptosis in uninfected , but rarely in infected , cells [43] . It can be speculated that hantavirus-infection may cause increased levels of bystander apoptosis in surrounding uninfected cells also in patients , and that this may account for the observed increased epithelial cell apoptosis . If this phenomenon also occurs in endothelial cells , and if it is involved in causing the increased vascular permeability during HFRS/HCPS , remains to be investigated . Vascular leakage is a hallmark of hemorrhagic fevers [3] . However , the mechanisms behind hantavirus infection-associated increased vascular permeability and clinical symptoms of HFRS/HCPS are not well understood . As eluded to above , a matter of discussion has been whether hantaviruses directly are involved in the pathogenesis and/or if immunologic components , i . e . , cytotoxic lymphocytes , are responsible [1] , [4]–[6] . In the latter case , both their cytotoxic as well as their cytokine producing capacity have been suggested to be associated with pathogenesis . Given the remarkably strong cytotoxic lymphocyte-responses of hantavirus-infected patients , with up to 50% of their peripheral blood CD8 T cells responding during the acute phase of infection and a unique NK cell-profile including highly elevated levels of activated NK cells present up to 60 days after symptom debut in many of the patients [7]–[9] , it has been suggested that cytotoxic lymphocytes are directly involved in the HFRS/HCPS-pathogenesis by causing increased vascular permeability by killing hantavirus-infected endothelial cells [1] , [6] . This hypothesis suggests that areas in the endothelial-cell barrier damaged by cytotoxic lymphocyte-mediated killing of hantavirus-infected cells are not efficiently repaired , as there is a shortage of thrombocytes because of the acute thrombocytopenia present in patients , which leads to the observed increased vascular permeability [1] , [6] . However , the present finding that hantaviruses efficiently inhibit cytotoxic lymphocyte-mediated killing of infected cells , together with reports that infected endothelial cells in patients are not damaged [4] , [10]–[12] , and the demonstration that CD8 T cells are not involved in ANDV-induced pathogenesis in a Syrian hamster model [44] , suggests that HFRS/HCPS-pathogenesis is not directly caused by cytotoxic lymphocyte-mediated killing of infected cells . On the other hand , it is possible that other cytotoxic lymphocyte functions are involved in HFRS/HCPS-pathogenesis , e . g . , their strong ability to secrete inflammatory cytokines that might cause increased vascular permeability by inducing gaps in the endothelial cell layer [6] . Although hantaviruses have no direct cytopathic effects on infected cells [1] , [2] , [4] , a direct role for hantaviruses in causing increased vascular permeability without damaging the infected endothelial cells has also been suggested [4] . Support for such reasoning comes from findings that pathogenic hantaviruses use the β3 integrin as receptor for cellular entry , and that hantavirus-β3 integrin interactions might contribute to increased vascular permeability [4] . Furthermore , hantaviruses induce VEGF-production [45] and sensitize infected cells to the effects of VEGF [46] , suggesting a role for VEGF in causing increased endothelial cell permeability . The finding that hantavirus nucleocapsid protein is a granzyme B and caspase 3-inhibitor together with the fact that hantaviruses primarily infect endothelial cells ( i . e . , cells that per se do not express granzyme B ) , suggest that hantaviruses have acquired a specific strategy to prevent cytotoxic lymphocytes from inducing apoptosis in infected endothelial cells . The present findings provide a plausible explanation as to why hantavirus-infected cells readily do not undergo apoptosis in patients , despite the apparently strong cytotoxic lymphocyte responses observed in infected individuals [7]–[12] . It is tempting to speculate that the extraordinary strong cytotoxic lymphocyte responses observed during , and long after , the acute phases of HFRS and HCPS [7]–[9] , [40] , [47] are in part caused by the inability of effector cells to eliminate hantavirus-infected endothelial cells .
Primary human umbilical vein endothelial cells ( HUVECs ) were grown according to the manufacturer's ( Lonza ) instructions using EGM-2 BulletKit ( Lonza ) . Before infection and co-culture experiments with NK cells , HUVECs were grown without supplementing the EGM-2 medium with hydrocortisone . The human lung epithelial cell line A549 ( American Type Culture Collection [ATCC] CLL-185 ) was grown in MEM supplemented with 5% FBS , 100 U/mL of penicillin , and 100 µg/mL of streptomycin ( all from Invitrogen ) . K562 cells ( ATCC CCL-243 ) were grown in complete medium ( RPMI 1640 , Invitrogen , supplemented with 10% FBS , 100 µg/mL L-glutamin , 100 U/mL penicillin , and 100 µg/mL streptomycin ) . Buffy coats from healthy human blood donors were obtained from the blood bank at the Karolinska University Hospital Huddinge , Stockholm , Sweden . PBMC were isolated by density centrifugation ( Ficoll-Hypaque; GE Healthcare ) followed by NK cell isolation ( Miltenyi ) . NK cells were cultured over night in complete medium supplemented with IL-2 ( 500 U/mL; Proleukin , Chiron Corporation ) . ANDV and HTNV were propagated on Vero E6 cells ( ATCC Vero C1008 ) as previously described [48] . For immune fluorescence experiments of cells exposed to staurosporine , cells were infected with multiplicity of infection ( MOI ) 0 . 01 , resulting in approximately 20% of cells being infected three days post infection . For all other experiments cells were infected with MOI 1; as earlier shown [9] this results in more than 80% of cells being infected at day three post infection ( Figure S3 ) . Cells were infected for 3 to 4 days before samples were collected for analyzes . Staurosporine and the pan caspase inhibitor Z-VAD-fmk were from BioVision while the caspase 3-inhibitor DEVD-CHO was from Sigma; all were used at a concentration of 2 µM each . Recombinant ANDV nucleocapsid protein [49] and rDHFR [50] were prepared as described [49] , [50] and subsequently subjected to dialysis against PBS . Recombinant active human caspase 3 and granzyme B were from BD Biosciences and Biovision , respectively . The monoclonal antibodies ( mAbs ) 7A2/D5 , 7B3/F7 and 1C12 , specific for hantavirus nucleocapsid protein , were used as previously described [49] , [51] . MAbs specific for PARP , caspase 3 and calnexin were all from Cell Signaling Technology . Overnight IL-2 stimulated NK cells and infected/uninfected HUVECs were co-incubated at an effector to target ratio of 1∶1 , or as specified in the text , at 37°C for 2 hours . Prior to the co-incubation , HLA class I on target cells was blocked using a combination of A6–136 hybridoma ( kindly provided by Dr D Pende , Istituto Nazionale per la Ricerca sul Cancro , Genoa , Italy ) and Dx17 mAbs ( BD Biosciences ) for 30 minutes at room temperature . After co-incubation with target endothelial cells , NK cells were stained with the following antibodies: anti-CD56 PE-Cy7 , anti-CD14 V500 , anti-CD19 V500 , anti-CD107a FITC ( all BD Biosciences ) , and anti-CD3 ECD ( Beckman Coulter ) . A dead cell marker ( DCM ) was used to exclude dead cells ( LIVE/DEAD Fixable Aqua Dead Cell Stain Kit , Invitrogen ) . Staining was performed for 30 minutes on ice in the dark in PBS containing 2% FBS ( FACS buffer ) , followed by 2 washes with FACS buffer , and subsequent fixation in PBS with 1% PFA for 30 minutes . Cells were acquired on a LSR Fortessa ( BD Biosciences ) and analyzed with FlowJo software version 9 ( Treestar ) . For the analysis , CD107a expression was evaluated on live NK cells ( CD56+CD3−CD14−CD19−DCM− ) . Uninfected and HTNV-infected endothelial cells , and K562 cells , were stained with CFSE ( Invitrogen ) following the manufacturer's instructions . IL-2 activated NK cells were incubated at different effector to target ratios with 20 , 000 target endothelial cells or K562 cells ( positive control ) for 4 hours at 37°C . To analyze NK cell-mediated target cell killing LIVE/DEAD Fixable Aqua Dead Cell Stain Kit was used to stain killed cells: DCM-staining was performed for 30 minutes on ice in the dark in FACS buffer , followed by 2 washes with FACS buffer , and subsequent fixation in PBS with 1% PFA for 30 minutes . Cells were acquired on a LSR Fortessa and analyzed with FlowJo software version 9 . For the analysis to detect specific target cells' death , the percentage of DCM-positive cells within CFSE-positive cells was evaluated . TUNEL assay ( Roche ) was performed as previously described [52] . After TUNEL reaction , cells were subjected to staining with anti-nucleocapsid protein mAb followed by FITC-conjugated goat anti-mouse IgG ( Sigma-Aldrich ) . Nuclei were counter stained with DAPI ( Sigma-Aldrich ) . Samples to be analyzed were mixed 4∶1 with NuPAGE 4× LDS sample preparation buffer ( Invitrogen ) , supplemented with 2 . 5% 2-mercaptoethanol , incubated at 96°C for 10 minutes , resolved on 10% NuPAGE Novex Bis-Tris gel ( Invitrogen ) and transferred to PVDF membranes . Blocking was performed at 4°C for 1 hour in PBS supplemented with 5% nonfat dry milk and 0 . 2% Tween 20 . The membranes were subsequently incubated with mAb for 1 hour at room temperature , followed by the addition of horseradish peroxidase-conjugated anti-mouse IgG ( Bio-Rad ) . Chemiluminescence substrate ( ECL Plus Western blotting detection kit , GE Healthcare ) was used following the manufacturer's protocol . Membranes were stripped in stripping buffer ( 10 mM 2-mercaptoethanol , 2% SDS and 62 . 5 mM TRIS-HCL [pH 6 . 7] ) . Samples from staurosporine-treated cells were homogenized in lysis buffer ( 150 mM NaCl , 2 mM EDTA , 1% NP-40 , and 50 mM Tris [pH 7 . 6] ) supplemented with complete protease inhibitor cocktail minitablets ( Roche Diagnostics ) prior to analyzes . Band densitometry was analyzed with the program Image J ( NIH ) . Caspase 3 and granzyme B activities were analyzed using specific activity assays according to the manufacturers' instructions ( Sigma for caspase 3 , Calbiochem for granzyme B ) . When analyzing cellular caspase 3 activity after staurosporine-treatment , total concentrations of cellular protein in samples were analyzed by a Bradford assay ( Biorad ) according to the manufacturers' instructions as internal control . To analyze nucleocapsid protein-specific inhibition of caspase 3 and granzyme B activity , recombinant human caspase 3 ( 0 . 1 µg ) or active recombinant granzyme B ( 0 . 1 µg ) was incubated with 1 µg recombinant ANDV nucleocapsid protein or 1 µg rDHFR as control , for 30 minutes or as stated in the text . When analyzing the effect different concentrations of the nucleocapsid protein had on caspase 3 and granzyme B activity , rDHFR were added to the samples to adjust the total amount of rANDV nucleocapsid protein+rDHFR proteins in all samples to 1 µg . Cellular caspase 3-activity was also analyzed using Fluorescent-Labeled Inhibitors of Caspases ( FLICA ) according to the manufacturer's ( Immunochemistry Technologies ) instructions . Briefly , target endothelial cells were co-incubated with IL-2-activated NK cells at an effector to target ratio of 1∶1 at 37°C for two hours , where the FLICA-probe FAM-DEVD-fmk was added for the last hour . Cells were subsequently stained with DCM , acquired on a LSR Fortessa , and analyzed with FlowJo software version 9 . Specific NK cell-induced activation of caspase 3-activity in target cells was calculated as percent of total active caspase 3-positive target cells minus percent of active caspase 3 positive endothelial cells not exposed to NK cells . Heat-inactivated ANDV ( 10 minutes at 96°C ) was incubated with either active recombinant human caspase 3 or granzyme B in reaction buffer ( BioVision ) at 37°C overnight and then subjected to immunoblotting for analysis of cleaved nucleocapsid protein . Edman degradation was performed by Alphalyse , Denmark . Empty plasmid ( PCMV-bios ) , and plasmids expressing wild-type ANDV nucleocapsid protein ( PCMV-bios-ANDV-N-wt ) and Asp285 to Ala285 mutated ANDV nucleocapsid protein ( PCMV-bios-ANDV-mut , constructed by Genscript , USA ) were prepared with EndoFree plasmid maxi kit ( Qiagen ) and transfected into A549 cells using Lipofectamine LTX ( LifeSciences ) according to the manufacturer's instructions . All data were analyzed using Prism ( GraphPad Software ) . Values are represented as mean ± SEM . Statistical analysis were performed using one-way or two-way ANOVA , followed by posthoc tests , or by two-tailed Student's t test . P-values<0 . 05 were considered significant , where * p<0 . 05; ** p<0 . 01 and *** p<0 . 001 . Recombinant ANDV nucleocapsid protein used in this study: AY228237 . 1 Hantaviruses used in this study: ANDV; AF291702 . 1 , AF291703 , AF291704 . 5 , HTNV; M14626 , M14627 , X55901 .
|
Rodent-born hantaviruses cause two severe emerging diseases with high case-fatality rates in humans; hemorrhagic fever with renal syndrome ( HFRS ) in Eurasia and hantavirus cardio-pulmonary syndrome ( HCPS; also called hantavirus pulmonary syndrome ( HPS ) ) in the Americas . A hallmark of HFRS/HCPS is increased vascular permeability . While endothelial cells are the main targets for hantaviruses , infection per se is not lytic . Patients suffering from HFRS and HCPS show remarkable strong cytotoxic lymphocyte responses including high numbers of activated NK cells and antigen-specific CD8 T cells . Hence , it has been suggested that cytotoxic lymphocyte-mediated killing of hantavirus-infected endothelial cells might contribute to HFRS/HCPS-pathogenesis . Here , we show that hantaviruses protect infected endothelial cells from being killed by cytotoxic lymphocytes . Further , we also show that hantaviruses inhibit apoptosis in general . Hantaviruses are negative-stranded RNA viruses encoding four structural proteins . Interestingly , the nucleocapsid protein was shown to inhibit the enzymatic functions of both granzyme B and caspase 3 , two enzymes crucial for cytotoxic lymphocyte-mediated killing of virus-infected cells . Our study provides new insights into the interactions between hantaviruses , infected cells , and cytotoxic lymphocytes , and argues against a role for cytotoxic lymphocyte-mediated killing of virus-infected endothelial cells in causing HFRS/HCPS .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"infectious",
"diseases",
"zoonoses",
"hantavirus",
"immune",
"cells",
"clinical",
"immunology",
"nk",
"cells",
"hantavirus",
"pulmonary",
"syndrome",
"immunology",
"viral",
"diseases"
] |
2013
|
Hantavirus-infection Confers Resistance to Cytotoxic Lymphocyte-Mediated Apoptosis
|
Sand flies deliver Leishmania parasites to a host alongside salivary molecules that affect infection outcomes . Though some proteins are immunogenic and have potential as markers of vector exposure , their identity and vector specificity remain elusive . We screened human , dog , and fox sera from endemic areas of visceral leishmaniasis to identify potential markers of specific exposure to saliva of Lutzomyia longipalpis . Human and dog sera were further tested against additional sand fly species . Recombinant proteins of nine transcripts encoding secreted salivary molecules of Lu . longipalpis were produced , purified , and tested for antigenicity and specificity . Use of recombinant proteins corresponding to immunogenic molecules in Lu . longipalpis saliva identified LJM17 and LJM11 as potential markers of exposure . LJM17 was recognized by human , dog , and fox sera; LJM11 by humans and dogs . Notably , LJM17 and LJM11 were specifically recognized by humans exposed to Lu . longipalpis but not by individuals exposed to Lu . intermedia . Salivary recombinant proteins are of value as markers of vector exposure . In humans , LJM17 and LJM11 emerged as potential markers of specific exposure to Lu . longipalpis , the vector of Leishmania infantum chagasi in Latin America . In dogs , LJM17 , LJM11 , LJL13 , LJL23 , and LJL143 emerged as potential markers of sand fly exposure . Testing these recombinant proteins in large scale studies will validate their usefulness as specific markers of Lu . longipalpis exposure in humans and of sand fly exposure in dogs .
Sand fly salivary proteins play a major role in blood feeding and Leishmania transmission [1]–[3] . Exposure to sand fly salivary proteins induces both cellular immunity and specific antibodies [3] , [4] . A relationship between the level of specific antibodies to saliva , vector exposure and risk of contracting disease has been demonstrated for different vector-host models [5]–[9] . Production of antibodies against mosquito and tick saliva not only contributed to development of host allergic reactions but was strongly related to risk of disease development [5] , [10] . Similarly , in an endemic area of Senegal , production of antibodies against Anopheles gambiae salivary proteins was identified as an indicator of the risk of malaria [10] . This correlation was also observed for tick exposure , where antibody production against tick saliva was associated with self-reported tick exposure and Lyme disease [11] . Recently , saliva of Triatoma infestans was shown to be a potential marker for vector infestation in domestic animals [12] . Therefore , the detection of antibodies against the saliva of hematophagous insect vectors could be used as an indicator of vector exposure and in some instances as an indicator for risk of contracting disease . Previous work shows that humans and animals exposed to sand fly bites or immunized with saliva can develop antibodies that recognize specific salivary proteins [4] , [7] , [13]–[15] . In São Luis , an area of endemic visceral leishmaniasis ( VL ) in Maranhão , Brazil , the presence of anti-saliva antibodies in humans strongly correlated with protection and the development of anti-Leishmania delayed-type hypersensitivity response [7] . Furthermore , individuals that poorly recognized salivary proteins developed anti-Leishmania antibodies associated with disease progression [7] . In contrast , in areas endemic for cutaneous leishmaniasis ( CL ) —such as Canoa ( Bahia , Brazil ) and Sanliurfa ( Turkey ) —the presence of anti-saliva antibodies correlated with risk of contracting disease [16] , [17] . The presence of antibodies to sand fly salivary proteins has also been demonstrated in animal reservoirs of leishmaniasis . In canines , two sand fly salivary proteins were recognized by sera of infected dogs from an endemic VL area in Brazil [18] . Hostomska et al . [14] reported the presence of anti-saliva antibodies to six different sand fly proteins in dogs experimentally exposed to Lutzomyia longipalpis bites . Importantly , foxes captured in Teresina , an endemic VL area in Brazil , also showed high levels of anti-saliva antibodies , particularly to a 44-kDa salivary protein from Lu . longipalpis , suggesting exposure to bites of this vector [19] . Hence , vector salivary proteins also represent a potential tool as markers of exposure to important reservoirs of disease . Identification of the sand fly salivary proteins recognized by the mammalian host will not only increase our understanding of vector-host interactions but will also aid in developing new epidemiological tools to correlate host exposure to vector sand flies with immunity or susceptibility to leishmaniasis . It will also help identify potential reservoirs of Leishmania . Here we describe a practical functional transcriptomic approach for the identification of the Lu . longipalpis salivary proteins most recognized by humans and canids ( dogs and foxes ) using sera from São Luis and Teresina , endemic areas for VL in Brazil [15] , [20] .
Lu . longipalpis ( Jacobina strain ) were reared at LMVR , NIAID , USA; Lu . verrucarum ( Peru strain ) and Phlebotomus perniciosus ( Italy strain ) at WRAIR , USA; Lu . intermedia ( Corte de Pedra strain ) were obtained from CPqGM ( FIOCRUZ , Bahia , Brazil ) . Females were used for dissection of salivary glands 5–8 days post-eclosion; SGH was prepared as described elsewhere [21] . Briefly , salivary glands were dissected and stored in sterile PBS ( pH 7 . 4 ) at −70°C . To obtain the homogenate , salivary glands were disrupted by ultrasonication and the supernatant collected after centrifugation at 15 , 000g for 2 minutes . A total of 14 human sera from from a VL-endemic region in São Luis ( Maranhão , Brazil ) [15] and 6 from a CL-endemic region in Canoa ( Bahia , Brazil ) [22] were used in this study . Informed written consent was obtained from parents or legal guardians of minors . The project was approved by the institutional review board from the Federal University of Bahia ( 1993 ) and the Federal University of Maranhao ( 1996 ) . Dog and fox ( Cerdocyon thous ) sera ( total of 8 and 11 , respectively ) were from animals captured in a VL-endemic area around Teresina ( Piaui , Brazil ) [19] . Fox and dog studies were approved in 2000 by the Brazilian agency for protection of the wildlife ( IBAMA/PI ) and in 2005 by the Federal University of Piaui . Sera were also obtained from dogs ( total of 6 ) experimentally exposed to Lu . longipalpis [14] . Dog studies were approved by Bayer Health Care AG ( Leverkusen , Germany ) and handled in accordance with the European guidelines for animal husbandry . DNA was amplified by polymerase chain reaction ( PCR ) using a forward primer deduced from the amino-terminus and a reverse primer encoding a hexhistidine motif . PCR amplification conditions were: one hold of 94°C 5 min , two cycles of 94°C 30 s , 48°C 1 min , 72°C 1 min , 23 cycles of 94°C 30 s , 58°C 1 min , 72°C 1 min , and one hold of 72°C 7 min . The PCR product was cloned into the VR2001-TOPO vector and sequenced [23] . A plasmid encoding a distinct salivary protein ( 1 µg/µl ) was injected intradermally into female Swiss Webster mice in 10 µl , three times at two-week intervals to generate polyclonal antibodies for each of the nine selected candidates [23] . Recombinant proteins were produced by transfecting 293-F cells ( Invitrogen ) with plasmid following the manufacturer's recommendations . After 72 h , the supernatant was recovered , filtered and concentrated to 30 ml in an Amicon concentrator device ( Millipore ) in the presence of Buffer A ( 20 mM NaH2PO4 , 20 mM Na2HPO4 , pH 7 . 4 , 500 mM NaCl ) . A HiTrap chelating HP column ( GE Healthcare ) was charged with 5 ml of 0 . 1M Ni2SO4 . The concentrated protein was added to the HiTrap chelating HP column that was then connected to a Summit station HPLC system ( Dionex , Sunnyvale , CA ) consisting of a P680 HPLC pump and a PDA-100 detector . The column was equilibrated for 30 min with Buffer A at 1 ml/min . Elution conditions were: 0-5 min , 100% Buffer A; 5-15 min , a gradient of 0% to 100% Buffer B ( Buffer A+50 mM imidazole ) ; 15-20 min , a gradient of 0% C ( Buffer A+500 mM imidazole ) to 10% C ( 90% B ) ; 20-25 min , 90% B and 10% C; min 25-30 , a gradient of 10% C to 20% C ( 80% B ) ; 30-35 min , 80% B and 20% C; 35-40 min , a gradient of 20% C to 100%C; and 40-50 min , 100% C . Eluted proteins were detected at 280 nm and collected every minute on a 96-well microtiter plate using a Foxy 200 fraction collector ( Teledyne ISCO ) . Five-microliter aliquots of all fractions were blotted on nitrocellulose and blocked with TBS-tween 3% non-fat milk for 1 h and then incubated for 1 h with anti-saliva antibodies , washed , and incubated for 1 h with an anti-mouse IgG ( H+L ) alkaline phosphatase-conjugated secondary antibody ( Promega ) . Positive fractions were developed with Western Blue stabilized substrate for alkaline phosphatase ( Promega ) . Positive fractions were run on sodium dodecyl sulfate ( SDS-PAGE ) and silver stained using SilverQuest ( Invitrogen ) . Imidazole was removed from positive fractions by dialysis overnight against PBS , pH 7 . 4 . Salivary glands ( 40 pairs approximately equivalent to 40 µg total protein ) or soluble recombinant sand fly salivary proteins ( 20 µg ) were run on a 4–20% Tris-glycine gel or on a 4–12% NuPAGE gel . After transfer to a nitrocellulose membrane using the iBlot device ( Invitrogen ) , the membrane was blocked with 3% ( w/v ) nonfat dry milk in Tris-buffered saline ( TBS ) -0 . 05% Tween , pH 8 . 0 , overnight at 4°C . After washing with TBS-T , pH 8 . 0 , the membrane was placed on a mini-protean II multiscreen apparatus ( Bio-Rad , Hercules , CA ) , and different lanes were incubated with various sera ( 1∶80 dilution , human and dog sera; 1∶50 dilution , fox sera ) for 3 h at room temperature . After washing with TBS-T , pH 8 . 0 , three times for 5 min , the membrane was incubated with either anti-dog IgG ( H+L ) alkaline phosphate-conjugated antibody ( 1∶10 , 000 ) ( Jackson Immuno Research ) for 1 h at room temperature for dog and fox sera or with anti-human IgG alkaline phosphate-conjugated antibody ( 1∶8 , 000 ) ( Sigma ) for human sera . Membranes were developed by addition of Western Blue stabilized substrate for alkaline phosphatase ( Promega ) , and the reaction was stopped by washing the membrane with deionized water .
Lu . longipalpis salivary glands contain a large number of secreted proteins ( figure 1A ) . Fourteen human sera from individuals living in São Luis , an area where Lu . longipalpis predominates , recognized a considerable number of these proteins , mainly between 15 and 65 kDa ( figure 1B ) . Eight dog sera from Teresina recognized a large number of salivary proteins , many of the same size as human sera as well as several proteins of different sizes ( figure 1B ) . As for foxes , 11 sera collected in the same endemic area as dogs recognized only a few salivary proteins and only one strongly of approximately 50 kDa ( figure 1B ) . To determine the specificity of human and dog sera for Lu . longipalpis salivary proteins , we tested the most reactive human ( two ) and dog ( one ) sera against salivary proteins from other sand fly species including Lu . intermedia which transmits CL in South America [24] , Lu . verrucarum which transmits CL in Central and South America [25] , and Phlebotomus perniciosus which transmits VL in Mediterranean countries [26] . Human sera recognized multiple bands of Lu . longipalpis saliva ( figure 2A ) . One of the 2 human sera also recognized two salivary proteins from Lu . intermedia ( figure 2A ) . We cannot exclude the possibility that this individual was weakly exposed to Lu . intermedia bites , as this species is also present , albeit rare , in São Luis [20] or to other non-abundant species in the area . All tested sera recognized proteins between 28 and 50 kDa from Lu . verrucarum saliva and a protein of approximately 40 kDa from P . perniciosus saliva ( figure 2A ) ; both species do not overlap with Lu . longipalpis in their geographical distribution . Given that Lu . longipalpis and Lu . intermedia are sympatric in several areas of Brazil [27] we decided to further investigate the possibility of antibody cross-reactivity between these two species . To address this , we tested six human sera from Canoa ( an area in Brazil endemic for CL where Lu . intermedia predominates ) against Lu . longipalpis saliva . These sera did not recognize any of the salivary proteins of Lu . longipalpis but recognized those of Lu . intermedia ( figure 2B ) . Together , these results suggest an overall low level of cross-reactivity between Lu . longipalpis and Lu . intermedia salivary proteins . Because in an endemic area there is no control of the diversity and intensity of exposure of hosts to sand fly bites—both of which can influence antibody response [14]—we compared dogs from Teresina , where Lu . longipalpis is prevalent , with dogs experimentally exposed to Lu . longipalpis bites . Overall , the reactivity of sera from experimentally exposed dogs was considerably lower than that of dogs from Teresina . Both dogs from Teresina and experimentally exposed dogs recognized proteins between 15 to 65 kDa ( figure 3 ) . Both groups recognized multiple proteins in Lu . longipalpis saliva but also a few in the saliva of Lu . verrucarum and P . perniciosus . This , together with results from human sera , suggests that antibodies against these proteins may be cross reactive for these two species . Additionally , while proteins from 28 to 50 kDa from Lu . intermedia were recognized by sera of dogs from Teresina , only one protein was poorly recognized by sera from experimentally exposed dogs . Sera from foxes were also tested but showed no cross-reactivity with the other species ( data not shown ) . Nine abundant transcripts corresponding to the predicted molecular weight of the most antigenic salivary proteins recognized by human , dog , and fox sera within the range of 15 to 65 kDa ( figure 1 ) were selected for expression ( Table 1 ) . Figure 4 shows a flow diagram of the approach used to express and purify the nine chosen recombinant salivary proteins . Notably , the same DNA plasmid is used for recombinant protein expression and antibody production . Nine different salivary proteins were expressed and a high level of purification was achieved by HPLC . Purification of recombinant salivary protein LJM17 resulted in a well separated peak eluting at 35–40 min ( figure 4B ) . Aliquots of eluted fractions were recognized by sera of mice immunized with LJM17 DNA plasmid ( figure 4C ) . SDS-PAGE of positive fractions shows a single band of approximately 50 kDa ( figure 4D ) , the expected size predicted by the LJM17 transcript . Similar results were obtained with the other eight expressed proteins: LJM111 , LJM11 , LJL143 , LJL13 , LJL23 , LJM04 , LJL138 , and LJL11 ( data not shown ) . To determine whether the nine expressed salivary recombinant proteins were recognized by sera from humans , dogs , and foxes , we chose those that recognized a considerable number of proteins ( from total sand fly saliva ) with some degree of variability for further analysis by western blot . Of the nine recombinant proteins tested , LJM17 , a yellow-related protein of 45 kDa , was the only protein recognized by sera from the 3 different hosts ( figure 5; data for foxes not shown ) . LJM11 , a 43-kDa protein also of the yellow family of proteins , was recognized by human and dog sera , while a third yellow-related protein , LJM111 ( 43 kDa ) was only recognized by human sera ( figure 5 ) . LJL23 , LJL13 , and LJM04 proteins were recognized only by dog sera; LJL143 was recognized by dog sera and weakly recognized by human sera ( figure 5 ) . LJL11 and LJL138 were not recognized by any of the sera tested ( data not shown ) . To confirm the specificity of LJM17 and LJM11 as potential markers of Lu . longipalpis exposure , we tested human sera from São Luis and Canoa where Lu . longipalpis and Lu . intermedia predominate , respectively . Both LJM17 and LJM11 were recognized specifically by human sera from São Luis but not from Canoa ( figure 6 ) .
Among parasitic diseases , leishmaniasis has one of the most complex epidemiologies . There are numerous Leishmania species and some cause a wide range of clinical manifestations and involve a large number of proven and potential reservoir hosts . In most cases , each form of leishmaniasis is transmitted by a sand fly species that acts as a principal vector; a few Leishmania species have multiple vector species [28] . In addition , endemic areas of leishmaniasis support several sand fly species other than the vector species responsible for Leishmania transmission . Finding tools to measure exposure of humans and reservoir hosts to specific vectors would provide valuable information regarding their contribution to parasite transmission and would be useful for assessing the risk of contracting disease . Several studies have demonstrated that anti-saliva antibodies can be used to assess exposure of humans and other Leishmania hosts to sand fly bites and suggested that sand fly salivary proteins represent attractive targets for development of specific markers of vector exposure [4] , [14] , [16] . To date , none of the salivary proteins of sand flies have been characterized for their immunogenicity and specificity in mammalian hosts , an important prerequisite for their reliability as markers of exposure . In the present work , we developed a robust method for producing and purifying recombinant salivary proteins . This approach proved highly successful in obtaining pure recombinant proteins that retain recognition epitopes . The purity of the produced recombinant salivary proteins is demonstrated by the presence of a single protein band following SDS-PAGE and silver staining; this level of purity was obtained for all recombinant salivary proteins tested . Soluble and pure recombinant proteins are likely to be properly folded and to better resemble native proteins . This should improve the sensitivity of the detection , enhancing recognition of such proteins by test sera and increasing the specificity of a test by decreasing the chances of false negatives often caused by impurities in the preparation . It is worth noting that the tested salivary recombinant proteins demonstrated specific responses , as two ( LJL11 and LJL138 ) of the nine proteins were not recognized by any of the tested sera . The seven immunogenic recombinant salivary proteins were differentially recognized by human , dog , and fox sera , the three host species investigated . Some proteins displayed host-specific recognition , reinforcing the importance of testing potential salivary markers for exposure against a variety of hosts to determine their range of applicability . Recombinant proteins LJM17 and LJM11 were strongly recognized by human sera from São Luis , a VL endemic area where Lu . longipalpis predominates , representing 66 . 4% of captured sand flies [29] . Notably , human sera obtained from Canoa , a CL area where Lu . intermedia represents 94% of the sand fly population [22] , did not recognize LJM17 or LJM11 . Lu . longipalpis and Lu . intermedia are sympatric species in many endemic areas of Brazil , often representing the two most abundant sand fly species [27] . As such , LJM17 and LJM11 can be considered as potential specific markers of exposure to Lu . longipalpis in areas where other man-biting sand fly species are present in negligible numbers [20] , [24] , [27] . LJM17 and LJM11 should be tested for specificity against other man-biting sand fly species that are relatively abundant in endemic areas—such as Lutzomyia whitmani [29]—to expand their utility as specific markers of exposure to Lu . longipalpis . Indeed , the faint bands recognized in Lu . intermedia saliva against one reactive human serum from São Luis ( figure 2 ) may have been due to cross reactivity between salivary proteins of this species and those of Lu . whitmani , reported to constitute as much as 24% of the sand fly population in this area [29] . The absolute specificity of LJM17 and LJM11 for Lu . longipalpis exposure therefore requires further confirmation through studies that target more sand fly species . Serum samples from São Luis also recognized salivary proteins from P . perniciosus , a vector of VL in the Old World [30] . This is interesting , as sera from inhabitants of Sanliurfa , Turkey , where P . papatasi and P . sergenti—two established Old World CL vectors—are abundant , did not react with salivary proteins of Lu . longipalpis [16] . Both LJM17 and LJM11 were recognized by dog sera from Teresina , endemic for canine VL [19] . A recent survey of the sand fly population from this area showed that Lu . longipalpis represented 99 . 7% of the collection [20] . Different from work done previously , here we detected six different salivary proteins as potential specific markers for exposure to Lu . longipalpis by using recombinant proteins for dogs ( figure 5 ) . The dog-biting status of other Lutzomyia species needs to be established before the specificity of these proteins as markers of exposure to Lu . longipalpis can be validated . However , this does not detract from their usefulness as potential markers for sand fly exposure for the evaluation of intervention studies in dogs . Although we did not test potential cross-reactivity of LJM17 and LJM11 with salivary proteins from other common vectors such as mosquitoes , kissing bugs , and black flies , extensive comparative transcriptomic analysis confirm that these two proteins are unique and distinct from those in the saliva of other arthropod vectors [31]–[35] . In conclusion , we have identified two salivary proteins from Lu . longipalpis , LJM17 and LJM11 , that were specifically recognized by sera from humans living in an endemic area of VL . Once tested on a wider scale , these proteins could become an important tool for accurate surveillance of this important vector of VL in Latin America .
|
Leishmania parasites are transmitted by the bite of an infected vector sand fly that injects salivary molecules into the host skin during feeding . Certain salivary molecules can produce antibodies and can be used as an indicator of exposure to a vector sand fly and potentially the disease it transmits . Here we identified potential markers of specific exposure to the sand fly Lutzomyia longipalpis , the vector of visceral leishmaniasis in Latin America . Initially , we determined which of the salivary proteins produce antibodies in humans , dogs , and foxes from areas endemic for the disease . To identify potential specific markers of vector exposure , we produced nine different recombinant salivary proteins from Lu . longipalpis and tested for their recognition by individuals exposed to another human-biting sand fly , Lu . intermedia , that transmits cutaneous leishmaniasis and commonly occurs in the same endemic areas as Lu . longipalpis . Two of the nine salivary proteins were recognized only by humans exposed to Lu . longipalpis , suggesting they are immunogenic proteins and may be useful in epidemiological studies . The identification of specific salivary proteins as potential markers of exposure to vector sand flies will increase our understanding of vector–human interaction , bring new insights to vector control , and in some instances act as an indicator for risk of acquiring disease .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"immunology/immune",
"response",
"public",
"health",
"and",
"epidemiology/screening"
] |
2010
|
Discovery of Markers of Exposure Specific to Bites of Lutzomyia longipalpis, the Vector of Leishmania infantum chagasi in Latin America
|
Dynamics of biomolecular assemblies offer invaluable insights into their functional mechanisms . For extremely large biomolecular systems , such as HIV-1 capsid that has nearly 5 millions atoms , obtaining its normal mode dynamics using even coarse-grained models can be a challenging task . In this work , we have successfully carried out a normal mode analysis of an entire HIV-1 capsid in full all-atom details . This is made possible through our newly developed BOSE ( Block of Selected Elasticity ) model that is founded on the principle of resonance discovered in our recent work . The resonance principle makes it possible to most efficiently compute the vibrations of a whole capsid at any given frequency by projecting the motions of component capsomeres into a narrow subspace . We have conducted also assessments of the quality of the BOSE modes by comparing them with benchmark modes obtained directly from the original Hessian matrix . Our all-atom normal mode dynamics study of the HIV-1 capsid reveals the dynamic role of the pentamers in stabilizing the capsid structure and is in agreement with experimental findings that suggest capsid disassembly and uncoating start when the pentamers become destabilized . Our results on the dynamics of hexamer pores suggest that nucleotide transport should take place mostly at hexamers near pentamers , especially at the larger hemispherical end .
The recent breakthroughs in experimental technology for structure determination , especially in single-particle cryo-electron microscopy [2] , have helped unveil many large structure assemblies at near atomic resolution for the first time . It is well recognized that a thorough knowledge of their dynamics can offer invaluable insight into their functional mechanisms and yet at the same time the enormous size of these systems poses a significant challenge to the computational simulations and analysis of their dynamics . Large scale computations comprising of millions of atoms are considered as one of the key problems by National Science Foundation’s ( NSF ) Molecular and Cellular Biosciences ( MCB ) program https://www . nsf . gov/funding/pgm_summ . jsp ? pims_id=504858&org=MCB . Normal mode analysis ( NMA ) [3–5] is a powerful tool for studying the intrinsic dynamics of biological assemblies . Mathematically , the core of all NMA computations involves solving a generalized eigenvalue problem of the Hessian matrix and the mass matrix . For extremely large assemblies , the source of the challenge in running NMA is the size of the Hessian matrix , whose dimension is in the same order as the number of atoms in the system . Precisely , for a system with N atoms , the Hessian matrix is of dimension 3N × 3N . For large systems with millions of atoms , it would take an extremely large amount of memory just to store the whole Hessian matrix even if a sparse matrix is used . To address this problem , two types of approaches have been developed . One is to use special eigenvalue solvers such as ARPACK [6] or order N technique [7 , 8] that are designed to compute quickly a small number of eigenvalues and eigenvectors . Similar to standard eigenvalue solvers , this type of approaches still require knowledge of a full Hessian matrix ( in the sparse matrix format ) , which can become severely limiting when dealing with extremely large systems such as HIV-1 capsid that has nearly 5 million atoms . The advantage of this type of approaches is that the accuracy is fully maintained and not compromised in any way . The other type of approaches for solving the eigenvalue problem of extremely large systems is by projection . RTB [9] and BNM [10] are two well-known approaches of this kind . Lezon and co-workers [11] , for example , successfully applied an RTB-based approach to compute the normal mode dynamics of HIV-1 capsid at a coarse-grained level . The advantage of projection-based methods is clear: it greatly reduces the size of the Hessian matrix . A major drawback of projection-based methods is the loss of accuracy , especially in normal modes of higher frequencies . In our most recent work [1] , we discovered a physical phenomenon that makes it possible to develop a new projection-based method that maintains all the advantages of projection-based methods and yet loses no or little accuracy . We discovered that the normal mode of a whole capsid at any given frequency ω is contributed nearly solely by vibrations of its individual capsomeres at around the same frequency , i . e . , there is a sharp resonance between the vibrations of a whole capsid and those of its capsomeres . Based on these observations , we were able to define a projection matrix P ( i ) ( 1 ≤ i ≤ m , where m is the number of capsomeres ) for each capsomere using the normal modes of the capsomere at a selected range of frequencies . The selection could be the modes below or a band of modes around a certain frequency [1] . For example , if there are N atoms in each capsomere and k modes are selected to represent P ( i ) , P ( i ) will be a 3N × k matrix whose columns are the selected modes . Given P ( i ) s , the projection matrix for the whole capsid is constructed as follows [1]: P = ( P ( 1 ) 0 ⋯ 0 0 P ( 2 ) 0 ⋮ ⋱ ⋮ 0 0 ⋯ P ( m ) ) , ( 1 ) and the projected Hessian matrix is [1]: H s = P ⊤ H P , ( 2 ) where H is the original Hessian matrix . Hs is now a much smaller matrix than H ( assuming that k ≪ 3N ) and thus is much easier to solve . The present work is a continuation of our previous work on resonance and focuses on the following issues that were not addressed in the resonance paper [1] . Specifically , 1 ) we conduct a quantitative assessment of the quality of the modes produced by the BOSE model . The assessment is carried out using four capsid test cases whose normal modes ( the benchmark ) can be obtained directly from the original Hessian matrix . In addition to “cumulative overlap” that is commonly used to assess mode quality , we develop also a new measure called “degeneracy-based overlap” for assessing the quality of modes . 2 ) We develop an additional measure that can be used to predict the quality of modes for the case when benchmark modes are not available , which is often the case and is the very reason for the existence of projection-based methods . 3 ) We address the issue of block selection and its effect on the performance of the BOSE model . This is especially relevant for capsids of which the composition of capsomeres , which are used as panel blocks in BOSE , is not obvious from the literature . In such a situation , we show that the aforementioned mode quality predicting measure can be used to determine what is the best choice for panel blocks . 4 ) Lastly , we perform for the first time an all-atom normal mode analysis of an entire HIV-1 capsid , a system with nearly 5 million atoms .
Recall that our aim here is to efficiently and accurately determine the normal modes of extremely large systems that have millions of atoms or more . We will focus on the low frequency normal modes in this work . The same method can be applied to obtain normal modes at other frequency ranges as well . The key realization behind the BOSE model is that large structure assemblies are made up of many components , or copies of proteins of the same or similar structures . BOSE reduces the complexity of the normal mode computation by effectively modeling the elasticity of each block with a small , selected number of normal modes . For the sake of simplicity , we assume in the following that the system being studied is composed of identical protein chains , even though the method still works otherwise . To evaluate the quality of normal modes determined from the projected Hessian matrix , either that of BOSE or of RTB , the following two measurements are used . Both of them require a comparison with the benchmark normal modes computed directly from the original Hessian matrix . In the following , we use v to denote a mode determined by a projection-based method ( BOSE or RTB ) , and p a mode determined from the original Hessian matrix ( the benchmark modes ) . The above mode quality assessment measures are still limited since in reality we generally don’t have the benchmark modes . The very reason for having the projection-based methods is that solving the eigenvalue problem of the original Hessian matrix is computationally prohibitive . Though we can assess the quality of BOSE modes on smaller systems for which benchmark modes are available and expect that the quality of BOSE modes remains the same by extrapolation , it is better to have a more direct way to predict the quality of modes . In our resonance paper [1] , we have shown that a capsid mode of frequency ω is contributed mostly by block modes at around the same frequency due to resonance . Consequently , to reproduce accurately a capsid mode of frequency ω , it is sufficient to include in the projection matrix only block modes of about the same frequency ( see Eq ( 4 ) ) . Therefore , our first major step to ensure the quality of BOSE modes is to use only BOSE modes whose frequencies are within the range defined by the block modes , as modes outside the frequency range are not of reliable quality due to the principle of resonance . Second , to quantify the cumulative contribution of a group of block modes to a given capsid mode v ˜ i , we define block-mode cumulative square overlap ( bmCSO ) as follows: bmCSO ( v ˜ i , t ) = ∑ j = 1 t ( ∑ k = 1 m c j , k ( i ) 2 ) , ( 15 ) where c j , k ( i ) is from Eq ( 10 ) . The inner summation ∑ k = 1 m c j , k ( i ) 2 represents the contribution to v ˜ i from the jth modes of all the m panel blocks in the system . Summation ∑ j = 1 t denotes the cumulative contribution of the first t modes of all panel blocks . Clearly , bmCSO ( v ˜ i , l ) = 1 accordingly to Eq ( 11 ) . Later on in Results section , we will show that bmCSO strongly correlates with d-overlap and thus can be used as a predictor of the quality of BOSE modes . bmCSO is a variant of cumulative square overlap ( CSO ) that was used in [17] . As aforementioned , capsomeres are the natural choice for panel blocks to be used in BOSE . The capsomeres often take the form of hexamers , pentamers , trimers , or dimers . For most capsids , the composition of the capsomeres is clear . Most capsomeres are so stable that they exist in isolation . For a few other capsids , it is not clear even from the literature what is the composition of the capsomeres: are they pentamers , trimers , or dimers ? Fortunately , as will be shown in Results section , our mode quality assessment measure is capable of indicating what is the best choice for panel blocks , especially when it is not obvious . In this work , we use four small capsids for benchmark tests before applying the BOSE model to the large HIV-1 capsid . The benchmark structures are prepared in the following way: The HIV-1 capsid structure is prepared by taking steps 1 , 3 , and 4 . The structures used by ANM are obtained by keeping only the Cα atoms . In our experiments , spring-based Normal Mode Analysis ( sbNMA ) [20] is used in all normal mode computations .
We use four capsids as test cases to evaluate the quality of BOSE modes . The four capsids are: capsid of Satellite Tobacco Necrosis Virus ( STNV , pdb-id: 4V4M ) [31] , capsid of Sesbania mosaic virus ( SeMV , pdb-id: 4Y5Z ) [32] , a mutant structure of the capsid of Grouper nervous necrosis virus ( GNNV , pdb-id: 4RFT ) [33] , and capsid of a lumazine synthase from the thermophilic bacterium Aquifex aeolicus ( AaLS , pdb-id: 5MPP ) [34] . The four capsids all have icosahedral symmetry . BOSE and RTB are both projection-based methods that reduce the size of the original Hessian matrix through restricting the motion space of structural building blocks , which can be either protein chains , capsomeres , or groups of residues . RTB restricts the motion of each building block to only rigid body motions . BOSE treats each building block still as a flexible unit , by modeling its elasticity using a selected subset of its normal modes . BOSE thus restricts the motions of each building block by allowing only vibrations within a certain frequency range . The allowed vibrations or normal modes define the selected elasticity of the block ( which is a capsomere ) . To have a fair comparison between BOSE and RTB , we let the two models have the same degrees of freedom for each capsomere . Consequently , the size of their reduced Hessian matrices are the same . The accuracy of either model is measured by comparing its modes with the benchmark modes determined from the original Hessian matrix . In this section , we present ways to predict the quality of BOSE modes when benchmark modes are not available . This will be the case when applying BOSE to compute the normal modes of new capsids , especially those that are so large that it is infeasible to compute their normal modes without employing a projection-based method . For such systems , we cannot apply cumulative overlaps or degeneracy-based overlaps to assess the mode quality since the benchmark modes are not available . Fig 3 shows the cumulative contributions of block modes to the normal modes of two whole capsids: STNV ( pdb-id: 4V4M ) [31] and AaLS ( pdb-id: 5MPP ) [34] . In the figure , all the 1800 modes of the capsid are equally divided into nine groups . Each curve represents the average bmCSO ( see its definition in Methods section ) of the modes within that group . The solid line curves represent capsid modes whose frequencies are within the range of the frequencies of the block modes . For these modes , we have confidence of their quality due to the principle of resonance [1] . The remaining groups of modes whose frequencies are out of range are drawn in dashed lines . The quality of these modes are unreliable . The bmCSO plot can be used to predict the quality of BOSE modes . To demonstrate this , we plot in Fig 4 bmCSO and degeneracy-based overlap ( d-overlap ) for all four capsids . The d-overlap assesses the quality of modes at different frequencies ( abscissa axis ) . The red line shows the block-mode cumulative square overlap ( bmCSO ) when the first 90% of the block modes are used , or bmCSO ( 90% ) . The blue line shows the degeneracy-based overlap of the BOSE modes . A d-overlap value close to 1 means high quality . The black solid vertical line marks the frequency upper limit of the panel block modes . It marks a frequency threshold above which capsid modes are no longer of good quality due to resonance , as seen in the sharp drop in the blue lines . Thus only modes below the frequency threshold ( i . e . , to the left of the vertical line ) are of interest . Under this frequency threshold , bmCSO ( 90% ) ( red line ) matches closely with d-overlap ( blue line ) , implying that both the frequency threshold and bmCSO ( 90% ) are good indicators of mode quality . Notice that in the zone between the dashed vertical line , which is 3 cm-1 to the left of the solid vertical line , and the solid vertical line itself , bmCSO ( 90% ) starts to drop significantly while d-overlap remains fairly high . The reason is that a capsid mode of frequency ω is contributed mostly by block modes of frequencies [ω − Δω , ω + Δω] where Δω is 3 cm-1 according to resonance [1] . The gap between the solid and dashed lines thus represents a “twilight” zone: the quality of modes in this frequency range is still fairly good according to d-overlap though it is not evident from the bmCSO ( 90% ) measure . In summary , when computing the normal modes of a new capsid using BOSE , we have two ways to assess the quality of the modes according to Fig 4 . One is to simply use the frequency threshold , the vertical solid line in Fig 4: normal modes below this frequency all have a high d-overlap value . The other is to use bmCSO ( 90% ) : normal modes with a large bmCSO ( 90% ) value also have a high d-overlap value . We investigate also the effect of panel block selection on the mode quality . This is especially necessary for cases when the choice of the capsomeres is not obvious . In the following , we consider two capsids: one is the capsid of Sesbania mosaic virus ( SeMV , pdb-id: 4Y5Z ) [32] and the other is a mutant structure of the capsid of Grouper nervous necrosis virus ( GNNV , pdb-id: 4RFT ) [33] . From the literature where the structures of these two capsids were first reported [32 , 33] , it is not entirely clear what are the capsomeres of these capsids: are they pentamers or trimers or something else ? In the following , for both capsids , two different choices of panels are tested: i ) using trimers as panel blocks and 90 modes per panel block; ii ) using pentamers as panel blocks and 150 modes per panel block . In both selection schemes , a total of 1 , 800 BOSE modes are generated for the whole system and compared with the benchmark modes of sbNMA . Fig 5 shows bmCSO and d-overlap plots of SeMV capsid when pentamers are used as panel blocks ( panels ( A ) and ( C ) ) and when trimers are used as panel blocks ( panels ( B ) and ( D ) ) . The mode quality is significantly better when pentamers are used . The same plots are repeated in Fig 6 for GNNV capsid , for which the opposite is true: the mode quality is significantly better when trimers are used as panel blocks . In both cases , even without seeing the d-overlap plots that shows the quality of the modes but cannot be computed without the benchmark modes , the bmCSO plots clearly reveal what panel choices are better . The right choice of panel blocks produces not only significantly higher bmCSO values but also more modes of reliable quality below the frequency threshold , i . e . , more solid lines and fewer dashed lines ( see panels ( A ) and ( B ) in Figs 5 and 6 ) . Table 1 lists the computational costs of BOSE , RTB , or sbNMA . The sbNMA Hessian matrix without symmetricity consideration would take more than 2 Tb memory space , which is too large for most computer systems . However , by applying group theory and taking advantage of the inherent icosahedral symmetry [35–37] , the Hessian matrix can be reduced to 10–15 Gb and normal modes can be obtained without losing any accuracy . On the other hand , both BOSE and RTB use a significantly less amount of memory . BOSE uses about 30% more computational time than RTB . The extra time is spent on computing the normal modes of the capsomeres . In this section , we apply the BOSE model to study the normal mode dynamics of an extremely large system in atomic details , the HIV-1 capsid . The HIV-1 capsid is a large structure with a molecular mass of 35 MDa and has nearly 5 million atoms . Because of its extremely large size , all-atom normal mode computations of this assembly are prohibitive on most computer systems . Our projection-based BOSE model allows us to perform all-atom normal mode computations of this large assembly for the first time . Our normal mode computations reveal in atomic details the intrinsic motion patterns of this large structure , particularly the dynamics of the pentamers , N-terminal loops of the capsid proteins , and hexamer pores .
Though the HIV-1 capsid studied in this work is one of the largest structures determined so far [12] , it is expected that atomic structures of even larger assemblies will come into light in the near future , such as structures of some of the bacterial microcompartments [43] , which are known to be made up of thousands of protein chains [43 , 44] . Faustovirus ( pdb-id: 5J7V ) [45] is another example . It has an astounding number , 8 , 280 to be precise , of chains . How can we ready ourselves for the dynamics studies of these giant assemblies ? A possible way to manage their immense size is to employ a hierarchical modeling of the whole structure . Specifically , a whole capsid may be first divided into fragments , with each fragment piece composed of a manageable number of capsomeres . Once this hierarchical structure is set up , one may apply the principle of resonance iteratively by obtaining first the dynamics of fragments from those of capsomeres and then the dynamics of the whole capsid from those of fragments . Such studies may help also pave the way for future simulations of organelles and even of cells . In this work , BOSE is applied solely to homomeric capsids . It is foreseeable that it can be easily extended to heteromeric capsids with minor adjustment and possibly , even to non-capsid assemblies . When a biomolecular system is composed of different proteins or even nucleic acids , different approaches may need to be combined . For example , when studying the ribosome that is made up of ribosomal RNAs and several dozens of distinct proteins , elastic units can be selected by considering the sizes of RNAs and proteins and their structural shapes , and the number of modes may be selected according to the size of each unit . We plan to extend BOSE to study such systems in future work . HIV-1 capsid uses the central pores of its hexamers to import nucleotides and to fuel encapsidated DNA synthesis [42] . The pores of the hexamers were thought to undergo an iris-like opening and closing motion [42] . Are the iris-like motions of the pores totally uncorrelated or fully synchronized somehow , or somewhere in between ? Currently little is known and it is certainly worth investigating . Note that HIV-1 capsid in this regard resembles closely bacterial microcompartments ( MCP ) , which also are enveloped by structural shells that are fully proteinaceous . The capsids of MCP serve as a diffusion barrier that isolates toxic reaction intermediates from the cytoplasm while allowing substrates , co-factors , and products to pass through [43 , 44] . The MCP capsids are composed of up to a few thousand shell proteins , most of which form hexamers or pseudo-hexamers ( trimers ) with central pores that are important functionally and are regulated dynamically . The dynamic regulation of MCP pores again is not well understood and normal mode analysis of these systems may provide the needed insights . It should be noted that at present the atomic structures of most MCP capsids are yet unknown except for a few , including the recent determined shell structure from Haliangium ochraceum [14] .
|
Supramolecular assemblies are large biomolecular complexes composed of hundreds or even thousands of protein chains . They function as molecular machines or as large containers that store or facilitate the chemical reactions of other molecules . Whatever they do , their functional mechanisms are tightly linked to their structures and intrinsic dynamics . Recently , due to breakthroughs in experimental techniques , many supramolecular assemblies have been determined , such as the capsid of human immunodeficiency virus ( HIV ) that is composed of nearly 5 millions of atoms . Computational studies of these systems are challenging due to their extremely large sizes . In this work , we have successfully carried out a dynamics study of an entire HIV capsid in full all-atom details . Our study reveals new insights into the dynamics of the N-terminal loops , the stabilizing role of the pentamers , and where the nucleotide transport may take place .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"resonance",
"frequency",
"medicine",
"and",
"health",
"sciences",
"classical",
"mechanics",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"vibration",
"pathogens",
"microbiology",
"viral",
"structure",
"organic",
"compounds",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"rna",
"viruses",
"basic",
"amino",
"acids",
"amino",
"acids",
"protein",
"structure",
"protein",
"structure",
"determination",
"proteins",
"medical",
"microbiology",
"hiv",
"resonance",
"microbial",
"pathogens",
"chemistry",
"viral",
"packaging",
"hiv-1",
"viral",
"replication",
"molecular",
"biology",
"physics",
"biochemistry",
"arginine",
"organic",
"chemistry",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"lentivirus",
"organisms",
"macromolecular",
"structure",
"analysis"
] |
2018
|
All-atom normal mode dynamics of HIV-1 capsid
|
CD95/Fas/APO-1 is a member of the death receptor family that triggers apoptotic and anti-apoptotic responses in particular , NF-κB . These responses are characterized by a strong heterogeneity within a population of cells . To determine how the cell decides between life and death we developed a computational model supported by imaging flow cytometry analysis of CD95 signaling . Here we show that CD95 stimulation leads to the induction of caspase and NF-κB pathways simultaneously in one cell . The related life/death decision strictly depends on cell-to-cell variability in the formation of the death-inducing complex ( DISC ) on one side ( extrinsic noise ) vs . stochastic gene expression of the NF-κB pathway on the other side ( intrinsic noise ) . Moreover , our analysis has uncovered that the stochasticity in apoptosis and NF-kB pathways leads not only to survival or death of a cell , but also causes a third type of response to CD95 stimulation that we termed ambivalent response . Cells in the ambivalent state can undergo cell death or survive which was subsequently validated by experiments . Taken together , we have uncovered how these two competing pathways control the fate of a cell , which in turn plays an important role for development of anti-cancer therapies .
Apoptosis is a program of cell death , which is essential for all multicellular organisms [1] . The crosstalk between apoptotic and anti-apoptotic pathways plays a key role in shaping life/death decisions in the cell . Furthermore , the success of anti-cancer therapies strongly depends on the efficiency of cell death induction in a single cell . However , a number of apoptotic stimuli are well known to activate strong anti-apoptotic response , which naturally might prevent apoptosis by upregulation of anti-apoptotic genes and , hence , counteract the effect of anti-cancer therapies [2 , 3] . In particular , members of the death receptor ( DR ) family have been reported to activate both apoptotic as well as anti-apoptotic responses [1 , 4] . The DR family is a subfamily of TNFR-superfamily , which includes TNF-R1 , CD95/Fas and TRAIL-R1/2 [5] . Triggering of the DR family with the cognate death ligands results in the formation of high molecular weight complexes , which induce cell death pathways and anti-apoptotic responses including activation of the transcription factor NF-κB [6–9] . The induction of a particular pathway in the DR system is a rather complex process , which is highly context dependent and multifactorial [8 , 10] . Importantly , molecular mechanisms underlying the intricate details of the cross-talk between apoptotic and anti-apoptotic pathways have not been established yet . The complexity of the response to DR activation stems from the fact that it is often heterogeneous within a population of cells [11–13] . The sources of the heterogeneity include genetic variations within cells , cell cycle effects , and stochastic effects from gene translation/transcription , which cumulatively might lead to different initial abundances of proteins within different cells [10 , 13] . For deciphering the multifactorial nature of cell death decisions in single cells , computational modeling paired with new experimental technologies providing a large number of data for the protein expression levels from individual cells are of indispensable value [12 , 14–16] . Of particular importance in this regard , is the cutting edge technology of imaging flow cytometry ( IFC ) , which combines microscopy and flow cytometry in one measurement enabling quantitative analysis of endogenous cellular protein levels estimated from a large number of cells simultaneously [17 , 18] . This feature of IFC is a strong advantage compared to confocal imaging that is mostly based on the application of artificially overexpressed activity probes , which might be often misleading . Furthermore , IFC in combination with machine learning provides a unique platform to quantitatively assign single cell events over a large number of cells and thereby has a strong advantage over confocal microscopy , which often needs manual analysis . In this study we addressed the interplay of apoptotic and anti-apoptotic pathways in single cells by analyzing the signaling network of the exemplified member of the DR family CD95 . CD95 stimulation leads to formation of the CD95 death-inducing signaling complex ( DISC ) [1] . The DISC comprises CD95 , the adaptor protein FADD , the initiator procaspase-8a/b ( p55/p53 ) , procaspase-10 and c-FLIP ( S1A Fig ) . After recruitment to the DISC , procaspase-8 builds death effector domain ( DED ) chains/filaments , formed via homotypic interactions between the DEDs of individual procaspase-8 molecules [19–21] . This provides the platform for homodimerization of procaspase-8 molecules , subsequent activation of procaspase-8 homodimers , and their processing with formation of the active caspase-8 heterotetramers p102-p182 ( S1A Fig ) . The activation of caspase-8 might be blocked by c-FLIP proteins , which incorporate into the DED chains and form heterodimers with procaspase-8 [21–23] . Interestingly , c-FLIP proteins act not only as inhibitors of CD95-induced apoptosis but are also essential for NF-κB activation [22 , 24] . Namely , the c-FLIP cleavage products p43-FLIP and p22-FLIP ( S1A Fig ) , have been reported to induce NF-κB [24–26] . The p43-FLIP cleavage product has been shown to play an essential role for CD95-mediated NF-κB activation [22] . The classical NF-κB activation pathway involves degradation of the NF-κB bound inhibitors of kappa B ( IκBs ) , which are phosphorylated by the IκB kinase ( IKK ) -complex containing two catalytic subunits ( IKKα and IKKβ ) and the regulatory subunit NF-κB essential modulator ( NEMO ) . Consequently , the NF-κB dimer ( p65/RelA and p50 ) is released to enter the nucleus , where activation of the transcription of the target genes takes place [27] . The intricate regulation of cell fate upon induction of CD95-mediated apoptosis vs . NF-κB at the single cell level remains largely unknown . Recently , a number of studies have highlighted the importance of the caspase-8 activation rate in single cells for the initiation of CD95-mediated apoptosis or survival [12 , 16] . However , the role of CD95-induced crosstalk of apoptosis with the anti-apoptotic signaling pathways such as NF-κB in the cell fate has so far only been addressed at the population level and never at the single cell level [24 , 28] . In this study we have investigated the regulation of these competing pathways in single cells using computational modeling , cutting edge technology of IFC and quantitative western blot . We show that cell-to-cell variability in the apoptotic phenotype does not result from NF-κB related stochasticity , but from heterogeneities in the composition of DED chains at the CD95 DISC .
Previously , it has been demonstrated that stimulation of CD95 with CD95L induces activation of both apoptotic and NF-κB pathways [6 , 9 , 24] . Several reports suggested that upon DR stimulation , in particular CD95 , some cells undergo apoptosis while other cells solely induce NF-κB [6] . Western blot analysis , as a “bulk population” measurement , does not allow to distinguish whether caspase activation and NF-κB induction occur in the same or distinct cells ( S1B Fig ) . To address this question , we used IFC , which allows to quantitatively analyze signaling events in a large number of single cells and to quantify these events at the population level . In particular , we monitored caspase-3 activation and p65 translocation to the nucleus and quantified them over 10 , 000 cells as described before [18] . HeLa cells overexpressing CD95 ( HeLa-CD95 cells ) were stimulated with 250 ng/ml CD95L followed by immunostaining with anti-p65 and anti-active caspase-3 antibodies in combination with staining of the nucleus with the DNA dye 7AAD ( Fig 1A and 1B ) . p65 translocation to the nucleus is a well described feature of NF-κB activation [29 , 30] . Accordingly , the similarity of the p65 and 7AAD signals in the nucleus serves as readout for NF-κB activation in IFC , which was also used in our study ( Fig 1B and 1C , S2A Fig ) . Furthermore , the timing of p65 translocation to the nucleus was consistent with a time course of p65 phosphorylation at Ser536 as well as degradation and phosphorylation of IκBα , which further verifies this approach of measuring NF-κB activation in single cells ( S1B and S1D Fig , S2B and S2C Fig ) . In accordance with our previous report [18] we have observed that CD95 stimulation results in the detection of four populations of cells: the first population does not show caspase-3 or p65 nuclear translocation . In the second one , only cells with p65-nuclear translocation while in the third population cells with active caspase-3 were observed . Finally , the fourth population consisted of cells that were characterized by activation of both pathways: apoptosis and NF-kB as manifested by both active caspase-3 and nuclear p65 detection . In this way , we observed that CD95 stimulation of HeLa-CD95 cells led to the appearance of nuclear p65 and active caspase-3 in the same cells ( Fig 1D ) . Furthermore , the response to CD95L stimulation was rather heterogenic: some cells showed stronger caspase-3 and NF-κB activation than others . This raised the question of how this heterogeneity manifests in life vs . death decisions and how it affects the dynamics of regulation of these two competing signaling pathways . To understand these processes and possibly delineate new molecular mechanisms beyond these distinct responses we applied computational modeling ( Fig 1E ) . The CD95 signal transduction network implemented in the model is initiated by the CD95 DISC leading to induction of apoptotic and NF-κB pathways [19 , 23] ( Fig 2 ) . After DISC assembly and DED chain formation , resulting in procaspase-8 activation and cleavage to p43/p41 , caspase-3 is activated , which leads to apoptosis ( Fig 2 ) . For this part we used a simplified version of the model introduced by Fricker and colleagues [31] . The topology of the anti-apoptotic pathway in the model was constructed in accordance with previous reports [24] . The c-FLIPL cleavage product p43-FLIP generated in the DED chain by procaspase-8 provides a link to the induction of the NF-κB pathway via activation of IKK complex that enforces phosphorylation and degradation of IκBα ( Fig 2 ) . IκBα binds NF-κB in the cytosol and thereby regulates the access of NF-κB to the nucleus . After degradation of IκBα , NF-κB enters the nucleus and induces expression of its target genes . The modeling of nuclear events following NF-κB activation was simplified . In our model only the synthesis of IκBα that keeps NF-κB in the cytosol , and the generation of a so-called generic negative regulator that deactivates IKKK and IKK , have been introduced . This part of the model strongly implements the topology of a previously introduced NF-κB model [32] . Taken together , in the constructed model active caspase-3 serves as a link between caspase-8 activation in the DED chain and apoptosis execution , while p43-FLIP generated in the DED chain connects the latter to the NF-κB pathway . In this way , the DED chain in our model gives rise to both caspase-3 and NF-κB activation promoting both pathways simultaneously . The DISC was modeled as a subnet and , subsequently , a submodel of the global CD95 network/model ( Fig 2 ) . The formation of three types of procaspase-8 homo- and heterodimers in the DED chain was introduced in the model: procaspase-8 homodimer , procaspase-8/c-FLIPL heterodimer , and procaspase-8/c-FLIPS heterodimer [23 , 31 , 33] . The first two dimers were catalytically active and underwent processing , while the procaspase-8 heterodimer with c-FLIPS was catalytically inactive ( Fig 2 ) . Furthermore , in order to simplify the modeling of the DED chain assembly , only ratios of DED proteins but not their variable positions in the chain were taken into account for the modeling . To define the ratio of procaspase-8/c-FLIP in the DED chain , their ratios were estimated in HeLa-CD95 cells via extrapolation from the corresponding ratios in SKW6 . 4 cells upon different stimulation strength ( Fig 3A ) . Consistent with previous reports , we also considered that strong stimulation results in shorter chains , while weak stimulation results in longer chains with a higher abundance of c-FLIP [20 , 22] ( Fig 3A ) . The model in SBML form is provided in the S1 Appendix . To account for the heterogeneity of the response to CD95 stimulation , e . g . the variabilities in the strength of caspase-3 vs . NF-κB activation , we used a stochastic modeling approach . Two sources of variability were incorporated into the model: extrinsic stochasticity , which was considered to arise from cell-to-cell variability in protein levels and intrinsic stochasticity resulting from different activities of the single components of intracellular signaling pathways arising from probabilistic processes of interactions between individual molecules . Since our knowledge regarding the initial distribution of all chemical species is limited to the proteins responsible for the DISC formation only procaspase-8 , c-FLIPL and c-FLIPS contribute to extrinsic noise in our model . We assume that these proteins obey a log-normal distribution with a standard deviation proportional to the mean . The proportionality factor was derived from data provided in [31] . It is known that intrinsic noise is very dominant when only a few copy numbers of a chemical species are present . Hence , the switching of genes from the active to the inactive state and the corresponding backward reactions are modeled as stochastic processes . In contrast the remaining reactions are treated deterministically . The generated model was trained against a merged set of single cell and population-level data , as indicated in the modeling workflow in Fig 1E . Single cell data were obtained via IFC analysis of caspase-3 activation and p65 nuclear translocation for stimulation strength from 10 to 250 ng/ml CD95L and for time intervals from 20 to 90 minutes ( Fig 3D , S3–S6 Figs , S1 Text ) . To broaden the data set for estimation of the model parameters , quantitative western blot analysis was performed for the analysis of signaling in HeLa-CD95 cells stimulated with 250 ng/ml CD95L ( Fig 3B and 3C , S3 and S4 Figs , S1 Text ) . This allowed measuring time-dependent processing of procaspase-8 to p43/p41 and p18 as well as the cleavage of c-FLIPL to p43-FLIP on the population level . Dynamic single cell and population data were integrated into the model by distinct approaches ( S1 Text ) . It turned out that the dynamics of the calibrated model are able to qualitatively capture the experimental data ( Fig 3C and 3D , S1 Text ) . We then used the calibrated model for subsequent analysis to compute the temporal evolution of single cell states via a hybrid version of the Gillespie algorithm accounting for the intrinsic noise [34] and Monte Carlo sampling of the initial conditions to compute extrinsic noise . To delineate the major features of the generated modeling network , we performed a detailed in silico analysis of central nodes of the model . The in silico investigation of the dependency of caspase-3 activation on the stimulation dose yields a two-dimensional landscape . In this landscape , upon increase of stimulation strength , means and standard deviations ( SDs ) of the active caspase-3 amounts rise exponentially ( Fig 4A ) . According to the model , the rise in SD occurs due to the extrinsic noise in the initial abundances of procaspase-8 and c-FLIP . The extrinsic noise leads to a high variability in DED chain configurations and consequently active caspase-8 amounts , resulting in different levels of active caspase-3 generated within distinct cells . For the high stimulation strength above 100 ng/ml CD95L a high level of active caspase-3 was predicted , which is in line with experimental observations by us and others [24 , 35] . Hence , model simulation indicates that in the range of 10 to 100 ng/ml CD95L the amount of active caspase-3 strongly depends on the stimulation dose , which is highly variable from cell to cell due to the extrinsic noise ( Fig 4A ) . The in silico analysis of p43-FLIP generation resulted in a rise of means and SDs in p43-FLIP for increasing stimulation dose and/or expression levels of c-FLIPL ( Fig 4B , S7A Fig ) . p43-FLIP is generated in DED chains , which implies that its generation is influenced by extrinsic noise . It turned out that consideration of extrinsic noise in silico did not alter the mean value of p43-FLIP generation ( S7B Fig ) . Instead extrinsic noise resulted solely in a spreading of the p43-FLIP distribution highlighted by an increased SD ( S7B Fig ) . Modeling predicted a relatively fast processing of c-FLIPL to p43-FLIP , accompanied by fluctuations of p43-FLIP with significant contributions of intrinsic and extrinsic noise . This is in accordance with a number of experimental data on fast processing of c-FLIPL to p43-FLIP in the DED chain [19 , 20 , 23 , 36] . The next question was how this behavior would influence the dynamics of NF-κB activation . Intriguingly , simulated time evolutions of NF-κB translocation to the nucleus have demonstrated a rather identical pattern for broad range of CD95L doses ( Fig 4C ) . In particular , the detailed dose-dependent analysis of CD95-induced NF-κB translocation to the nucleus determined several regions of NF-κB response to CD95 stimulation in silico ( Fig 4D ) . For stimulation doses between 0 . 5 and 5 ng/ml CD95L NF-κB activation took place , but there was a huge SD , while for stimulation doses above 5 ng/ml the activation was very robust against intrinsic noise , which is visualized by stable temporal patterns of means and SDs in a broad interval of stimulation strength ( Fig 4D ) . Intriguingly , in contrast to several NF-κB activation pathways [37–39] , different initial levels of nuclear p65 influence NF-κB activation did not show any dependence on NF-κB activation according to the analysis in silico ( Fig 4E–4G , S7C Fig ) . The dependency of NF-κB activation on the initial concentration of c-FLIPL at the DISC showed that for c-FLIPL concentrations below the endogenous level an activation of NF-κB took place , but it was characterized by a high level of noise due to the low abundance of nuclear NF-κB ( Fig 4F ) . For higher initial concentrations stable patterns of NF-κB activation were observed ( Fig 4F ) . Further , the analysis of the extrinsic variability of the DISC leading to p43-FLIP generation and consequently NF-κB activation , has demonstrated that extrinsic noise does not significantly influence the p65 translocation dynamics ( S7D Fig ) . Modeling uncovered a very intriguing feature of CD95-mediated NF-κB activation: a stable pattern of activation , which is not dependent on the stimulation dose . In this way , CD95-mediated NF-κB induction is fundamentally different from a number of other NF-κB activation pathways . Hence , CD95-induced apoptosis and NF-κB have two opposite types of behavior: caspase-3 variability increases for all stimulation strengths , while NF-κB activation shows a robust behavior . This leads to the suggestion that starting from dosages above 5 ng/ml CD95L , heterogeneity in the apoptotic response should be dictated by the variability in the DED chain formation at the DISC rather than the stochastic gene expression in the NF-κB response . Next , we aimed on deriving a corresponding parameter , which can be used for predicting cell fate . A cell undergoes apoptosis shortly after a critical concentration of caspase-3 is reached within a cell , which is known in the literature as the ‘point of no return’ [12 , 35 , 40] . The ‘point of no return’ was accordingly introduced into our model as an in silico value , which reflects the amount of active caspase-3 , that is required for the particular cell to undergo apoptosis ( Fig 5A ) . In order to estimate the critical amount of caspase-3 we used a quadratic discriminant analysis to discriminate between viable and apoptotic cells based on the caspase-3 fluorescence . This revealed a fluorescence threshold separating two subpopulations ( S8 Fig ) . By linking the fluorescence to the abundance of caspase-3 estimated by our mathematical model we were able to compute the critical amount of caspase-3 , which marks the ‘point of no return’ . The time when the ‘point of no return’ is reached strongly depends on the stimulation dose ( Fig 5A ) . NF-κB activation induces the transcription of genes that may counter the apoptotic pathway [41] . To delineate the connection between timing of NF-κB activation and apoptosis and thereby to get more insight into interplay between two pathways two new parameters were introduced: the time of decision ( TOD ) , which is the time interval from stimulation to the ‘point of no return’ and the time of survival signaling ( TOS ) , which is the interval from the maximum of NF-κB translocated to the nucleus to the ‘point of no return’ ( Fig 5A ) . TOD and TOS are governed by the stimulation dose , since caspase-3 activation is sensitive to dose variation ( Fig 5A ) . We hypothesized , that cells with a larger TOS/TOD ratio tend to survive the apoptotic stress , due to more time to counter the activation of the apoptotic machinery . In contrast , cells with small ratios fail to fight apoptosis due to the lack of time to synthesize anti-apoptotic proteins downstream of active NF-κB . For large doses the TOS/TOD ratio is negative , since the timing of the ‘point of no return’ precedes the timing of maximum nuclear NF-κB activation . In contrast , the TOS/TOD ratio tends to one for small stimulation doses since the maximum nuclear NF-κB activation occurs much earlier than the ‘point of no return’ . Hence , TOS is approximately equal to TOD and their ratio yields one . Next , we analyzed the effects of extrinsic noise from the DISC and intrinsic noise from NF-κB activation on the TOS/TOD ratio ( Fig 5B , dark blue line ) . Both effects jointly led to a distributed TOS/TOD ratio ( Fig 5B , blue region ) . In particular , NF-κB activation has its maximal variability from 0 . 5 to 5 ng/ml CD95L ( Fig 4D , right panel ) , where TOS/TOD variance is negligibly small ( Fig 5B ) . In the range of 5 to 10 ng/ml CD95L , NF-κB is activated but its variability is of minor importance ( Fig 4D , right panel ) . On the other hand , variability in the abundance of the active caspase-3 starts to play a role already from very small CD95L stimulation doses and therefore has a major influence on the variability of TOS/TOD ( Fig 5B ) . This in silico analysis pointed out that a response to CD95L stimulation between 5 and 10 ng/ml apparently presents a range of an ambivalent response , where single cells can either undergo apoptosis or survive upon the same stimulation strength . To more closely relate the TOS/TOD ratio from our model to experimental data on apoptosis induction , we aimed to derive a critical ratio rcrit of TOS/TOD , which could predict cell fate in a single cell . A cell with TOS/TOD ratio > rcrit is very likely to survive , whereas a cell with TOS/TOD ratio < rcrit tends to undergo apoptosis . We expect rcrit to be positive so that the maximum of nuclear NF-κB activation occurs before the ‘point of no return’ . In order to estimate the critical ratio we performed a least squares fit of the models using single cell simulations to match the dose-response curve from cell death data ( Fig 5C , dashed line ) , which led to the value of rcrit = 0 . 965 . The model qualitatively captures the sigmoid response , but there are discrepancies between theory and experiment . We believe that most likely this is due to additional biological variability , which was not accounted for in the model so far . If we assume that this variability causes slight variations of the critical TOS/TOD ratio in single cells we can fit our model almost perfectly to the experimental data , which strongly supports the validity of the TOS/TOD parameter for prediction of the life and death decisions ( Fig 5C ) . By a least squares fit we obtained a relative SD of 1 . 5% of its mean . More statistics regarding the TOS/TOD ratio can be found in the S1 Text . To further validate the TOS/TOD model we checked experimentally whether by influencing the TOS/TOD ratio apoptosis rates can be perturbed . For this purpose we used several CD95L stimulation strengths up to 50 ng/ml and longer stimulation periods of 15 hours in combination with inhibitors of caspases ( zVAD-fmk ) and NF-κB activation ( IKK inhibitor VII ) . The distribution between living and apoptotic cells was analyzed via caspase-3 intensity and morphological features of the cells [42] ( Fig 5D , S1 Text ) . The model predicted that the addition of the NF-κB inhibitor should significantly reduce TOS and thereby lead to a decrease of the TOS/TOD ratio compared to cells in which NF-κB is not inhibited . Addition of the IKK inhibitor indeed resulted in more apoptosis , which was detected upon low stimulation strengths of 5 ng/ml and 10 ng/ml ( Fig 5D ) . This further confirmed that stimulation with CD95L at this dose range induces ambivalent response . Accordingly , the addition of caspase inhibitor increases the TOS/TOD ratio via increase of both TOS and TOD and therefore blocks apoptosis , which was subsequently observed in these experiments ( Fig 5D ) . We also tested the effects of caspase and IKK inhibition using a different experimental approach based on live cell imaging and obtained similar results . Addition of IKK inhibitor VII resulted in a higher amount of apoptotic caspase 3/7 positive cells ( S9 Fig , S1–S3 Videos ) . Importantly , the most sensitive region to the perturbations of TOS/TOD turned out to be the region between approximately 5 and 10 ng/ml , which further confirmed the model predictions . Taken together this analysis uncovered three different regions of cell fate that can be identified with the TOS/TOD ratio: cell death , ambivalent response and cell survival , providing important insights into pathway dynamics and paves the way towards possible therapeutic applications .
A number of models addressing apoptosis and cell fate in single cells have been created [10 , 12 , 13] . However , single cell modeling of the interplay between apoptosis and anti-apoptosis pathways , which is a key feature of this model , has been missing so far . The understanding of this interplay at the single cell level is very important in the context of creation of efficient anti-cancer therapies specifically inducing apoptosis in every single cell . To understand the behavior of these competing pathways , we connected the strength of the apoptotic vs . NF-κB response with the stoichiometry of the DISC/DED chains using in silico analysis and single cell experimental data . Most of the models describing single cell behavior use experimental data that are generated with artificially overexpressed caspase activity probes or tagged NF-κB proteins that might be influencing activation of this pathway . In contrast , our study is entirely focused on the analysis of endogenous proteins at the single cell level which is deemed possible due to the state of the art high throughput IFC analysis . Furthermore , IFC provides a platform to quantitatively assign single cell events over a large number of single cells using machine learning [42] . The latter potentially provides an enormous asset for building computational models and our study presents one of the first models of apoptosis and anti-apoptosis pathways implementing IFC data . Among contemporary approaches there are several powerful single cell technologies that attract a major attention and allow to quantify endogenous proteins at the single cell level , in particular , IFC and single cell proteomics . Due to limited imaging channels in IFC this technology allows to follow only a limited number of proteins ( up to 10 ) and is limited to the availability of suitable antibodies and dyes . In contrast to this , single cell proteomic approaches allow following a higher number of proteins . The advantage of IFC is that it allows following selected proteins . Therefore , in particular , with respect to quantitative biology and modeling signaling pathways IFC is undoubtedly the technology of choice . In this study apoptosis and anti-apoptosis pathways in single cells upon DR stimulation were analyzed using a stochastic model . Strikingly , the model shows that NF-κB activation is highly variable around 1 ng/ml CD95L , whereas it becomes relatively robust at a higher CD95L stimulation strength starting from 5 ng/ml . In contrast , caspase-3 dynamics are highly variable for all dose ranges . Heterogeneity in caspase-3 activation is directly linked to the chain configuration at the DISC . The two distinct types of stochastic behavior of caspase-3 and NF-κB activation at a dose range of 5 to 10 ng/ml , lead to the fact that both scenarios of life or death are possible within this concentration range . This in turn creates a dose range of ‘ambivalent' response . Hence , we concluded that the major source of ambivalent response upon 5 to 10 ng/ml CD95L stimulation is heterogeneity in DED chain configuration , which results from extrinsic noise . To more closely link our mathematical model of caspase-3 and NF-κB dynamics to experimental data , we introduced a parameter that determines the life/death decision in the cell and the threshold for apoptosis , the TOS/TOD ratio . Using this parameter could give a plausible explanation of the ultimate apoptotic response mechanism and its dose-dependent variability . In particular , the TOS/TOD ratio provides a clear explanation to the events occurring in the ambivalent response region between 5 and 10 ng/ml . In particular , even though both pathways are active upon stimulation above 5 ng/ml CD95L , cells can ‘fight’ apoptosis , resulting from the possibility of late activation of apoptotic downstream targets ( TOS/TOD>rcrit ) . For dosages around 5 ng/ml CD95L the chain configuration may in rare cases induce a quick response in caspase-3 ( TOS/TOD<rcrit ) and the cell might undergo apoptosis ( Fig 5B ) . Interestingly , if the TOS/TOD curve were steeper ( Fig 5B ) , the ambivalent response range would decrease , whereas a more flat TOS/TOD curve increases the dose range of ambivalent response . We validated this behavior experimentally by inhibiting IKK and caspases , which fully confirmed TOS/TOD model predictions . From our results we further conclude that on a single cell level , a life/death decision is very much dependent on the current state of the cell and the level of expression of anti-apoptotic vs . apoptotic proteins . For high stimulation strengths , the apoptotic decision prevails , whereas low stimulation strengths can also result in an anti-apoptotic response . The latter fits well to the recent findings by Roux et al . who have demonstrated that for low rates of DR-induced caspase-8 activity the cells might survive , which might be connected to the induction of the NF-κB pathway as we show in this study [12] . Here , we reason that this can be caused by dose/timing effects of two key signaling molecules related to life ( NF-κB ) and death ( caspase-3 ) . From our computational model we observed that activation of NF-κB is insensitive to stimulation dose variations above 5 ng/ml CD95L . This stems from the fact that NF-κB activation has a binary character owing to a time scale separation effect ( signaling vs . gene expression time scale ) . Consequently , the stochastic gene expression related to NF-κB does not seem to be the main source of heterogeneity in the life/death decision . Rather , from our TOS/TOD ratio hypothesis , variations in the chain configuration and related caspase-3 fluctuations seem to be the dominant source of variability in the life/death response . Defining the optimal strength of cell death induction leading to the eradication of cancer cells is a key step in anti-tumor therapy . The dynamics of cell death induction is often neglected despite its very important role in defining the life/death outcome . Furthermore , as we can conclude from analyzing the dynamics of CD95 signaling , the cross-talk between apoptotic and non-apoptotic pathways has to be considered for predicting the outcome of DR stimulation . The future challenge will be to define the dynamics of cell death responses in cancer vs . normal cells supported by powerful computational modeling in order to develop efficient anti-tumor therapies .
Cervix carcinoma HeLa cell line with overexpression of CD95 ( HeLa-CD95 ) [24 , 28] were cultivated in 5% CO2 at 37°C in DMEM/Hams F12 media ( Biochrom , Berlin , Germany ) with addition of 10% heat-inactivated FCS ( Life Technologies , Darmstadt , Germany ) , 1% Penicillin/Streptomycin ( Merck Millipore , Darmstadt , Germany ) and 10 ng/ml Puromycin ( Sigma Aldrich , Taufkirchen , Germany ) . 5*105 HeLa CD95 cells were stimulated with CD95L for indicated time periods and concentrations [31] . After stimulation , cells were harvested by scraping , washed with PBS and lysed in lysis buffer ( 20 mM Tris HCl , pH 7 . 4 , 137 mM NaCl , 2 mM EDTA , 10% glycerol , 1% Triton X-100 , Protease Inhibitor mix ( Roche , Mannheim , Germany ) ) . 20 mg of protein lysate was separated on 12% stain-free SDS gels and blotted on nitrocellulose membrane ( both Biorad , Hercules , USA ) . Membranes were blocked with 5% non-fat dried milk in PBS-T ( 0 . 05% Tween 20 in PBS ) for one hour , washed three times with PBS-T and incubated with primary antibody overnight at 4°C . Before addition of the HRP-coupled isotype-specific secondary antibodies ( Santa Cruz Biotechnology , Dallas , USA ) the membrane was washed four times with PBS-T . After incubation the membrane was washed three times with PBS-T . HRP substrate ( Luminata forte , Merck Millipore ) was added and the chemiluminescence signals were captured with a ChemiDoc imaging system ( Biorad ) . The images were quantified with the ImageLab 5 . 1 Software ( Biorad ) . Special care was taken to avoid overexposure of images . Lanes and bands were selected manually and subsequently quantified . Anti-phospho-IκBα and anti-IκBα antibodies were purchased from Cell Signaling Technology ( Danvers , USA ) . Anti-actin antibody was purchased from Sigma Aldrich ( Taufkirchen , Germany ) . Anti-caspase-8 C15 and anti-c-FLIP antibodies were a kind gift of Peter H . Krammer ( DKFZ , Heidelberg , Germany ) . 5*105 HeLa cells were stimulated for indicated time periods with CD95L [33] . When indicated , cells were pre-stimulated with 50 μM zVAD-FMK ( Bachem AG , Bubendorf , Switzerland ) or 10 μM IKK inhibitor VII ( MerckMillipore ) for 30 minutes . After stimulation medium was removed , cells were washed with PBS and detached with trypsin-EDTA solution ( Life Technologies , Darmstadt , Germany ) for 5 minutes at 37°C . Afterwards , cells including medium , washing PBS , and trypsin were spun down at 500xg at 4°C for 5 minutes . Cell were fixed with 3% formaldehyde in PBS for 10 minutes at room temperature , permeabilized with 90% ice-cold methanol for 30 minutes on ice and washed twice with incubation buffer ( 5 g/l albumin fraction V ( AppliChem , Darmstadt , Germany ) in PBS ) . For antibody staining , cells were resuspended in 50 μl incubation buffer . Cells were stained for one hour in the dark with 1 μl anti-caspase-3 antibody , recognizing caspase-3 cleaved at Asp175 , conjugated to AlexaFluor488 or AlexaFluor647 , 0 . 5 μl anti-NF-κB-p65 antibody , conjugated to phycoerythrin ( PE ) , ( both purchased from Cell Signaling Technologies ) . After staining , cells were washed with incubation buffer and resuspended in 40 μl PBS . At least 5 minutes before measurement , 3 μl of the DNA dye 7AAD ( BioLegend , San Diego , USA ) was added . Imaging flow cytometry measurements were carried out using an Amnis FlowSight ( Amnis/MerckMillipore ) . 10 , 000 events per sample were acquired using the channels 1 ( bright field , 435–505 nm ) , 3 ( 560–595 nm ) , 5 ( 642–745 nm ) , 9 ( bright field , 560–595 nm ) and 11 ( 642–745 nm ) . The samples were excited with a 488 nm laser with 60 mW and a 642 nm laser with 100 mW laser power . Automated color compensation was performed with cells stained with one dye . Data were analyzed with IDEAS software version 6 . 1 ( Amnis/ MerckMillipore ) . Single cells , with staining for 7AAD and p65 were selected and analyzed for p65 translocation [18] . Fluorescence intensity for antibody staining was analyzed with the intensity function of IDEAS . On the day before stimulation 5*104 HeLa-CD95 cells per well were seeded into a flat-bottom 96-well plate . Before stimulation media was replaced with 100 μl of media including IncuCyte Caspase-3/7 Apoptosis Assay Reagent ( Essen Bioscience , Ann Arbor , USA ) according to the manufacturer’s instructions . Then cells were incubated for 30 minutes with 10 μM IKK inhibitor VII ( MerckMillipore ) or 50 μM zVAD-FMK ( Bachem AG ) followed by CD95L stimulation . The first images were acquired 30 minutes after stimulation with IncuCyte ZOOM system ( Essen Bioscience ) . Data analysis and videos were made with IncuCyte ZOOM software version 2015A1 ( Essen Bioscience ) . To perform single cell simulations , a mathematical model was developed and calibrated to experimental data . Since it was assumed that the formation of the DISC is several magnitudes faster than the downstream processes , the DISC formation was suggested to take place immediately after addition of CD95L . To account for the dose dependency of the DISC stoichiometry [20] the protein abundances of procaspase-8 , c-FLIPL , and c-FLIPS were modeled as dose dependent functions . We derived a working chain model by using data from Schleich et al . [20] . Although Schleich et al . used the SKW6 . 4 cell line and not Hela-CD95 cells , it was assumed that the same chain configuration prevails . Thus , the experimental data from [20] were extrapolated for this study . A complete list of species with initial conditions is provided in the S1 Text . Biological systems often exhibit stochastic behavior , especially in gene networks with low copy numbers [43] . Therefore , the reactions accounting for the switching of genes between active and inactive states were modeled as stochastic processes and the remaining reactions deterministically . For simulation a hybrid version of the stochastic simulation algorithm [34] was used . Extrinsic variability was modeled as distributed initial conditions ( including heterogeneous chain formations at the DISC ) , which were simulated with Monte Carlo sampling . Thus , we ended up with a hybrid stochastic-deterministic model . To efficiently simulate intrinsic and extrinsic noise during parameter optimization we used the sigma point method [44] combined with the hybrid stochastic simulation algorithm describing the interplay of distributed properties and low copy number effects at the gene expression level [45] . Since evolutionary algorithms haven been previously reported to yield excellent results when stochastic simulations are performed [42 , 46] we used a genetic algorithm for optimization . For model calibration , western blot and flow cytometry data were used . Further details on the computational model and parameter optimization can be found in the S1 Text and S1 Appendix .
|
Activation of death receptor ( DR ) family has been reported to activate both apoptotic as well as anti-apoptotic responses . Molecular mechanisms underlying the intricate details of this crosstalk have not been established yet . Here we show that these pathways are triggered simultaneously in one cell . Furthermore , using stochastic computational modeling we uncovered how an individual cell undergoes apoptosis , while other cells survive upon the same DR activation conditions . This was only possible by combination of computational modeling supported by experimental validation based on the state of the art single cell analysis . The latter included cutting edge technology of imaging flow cytometry , which combines microscopy and flow cytometry in one measurement circuit enabling quantitative analysis of endogenous cellular protein levels estimated from a large number of cells simultaneously . This allowed to shed the light on the question how a single cell possibly avoids apoptosis , which is a highly actual topic in the field of cancer research and development of efficient anti-cancer therapies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"flow",
"cytometry",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"dose",
"prediction",
"methods",
"cancer",
"treatment",
"cell",
"processes",
"simulation",
"and",
"modeling",
"oncology",
"pharmaceutics",
"immunoprecipitation",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"gene",
"expression",
"precipitation",
"techniques",
"spectrophotometry",
"cytophotometry",
"cell",
"staining",
"cell",
"biology",
"apoptosis",
"genetics",
"biology",
"and",
"life",
"sciences",
"spectrum",
"analysis",
"techniques"
] |
2018
|
Quantitative single cell analysis uncovers the life/death decision in CD95 network
|
There is a significant need for improved treatments for onchocerciasis and lymphatic filariasis , diseases caused by filarial worm infection . In particular , an agent able to selectively kill adult worms ( macrofilaricide ) would be expected to substantially augment the benefits of mass drug administration ( MDA ) with current microfilaricides , and to provide a solution to treatment of onchocerciasis / loiasis co-infection , where MDA is restricted . We have identified a novel macrofilaricidal agent , Tylosin A ( TylA ) , which acts by targeting the worm-symbiont Wolbachia bacterium . Chemical modification of TylA leads to improvements in anti-Wolbachia activity and oral pharmacokinetic properties; an optimized analog ( ABBV-4083 ) has been selected for clinical evaluation .
The filarial worm diseases onchocerciasis ( “river blindness” ) and lymphatic filariasis ( LF , “elephantiasis” ) , though not typically lethal , produce substantial morbidity , social stigma and loss of economic opportunity in tropical and subtropical regions throughout the globe [1 , 2] . Nearly 150 million people are currently infected with these parasites , with a greater number at risk; more than 40 million suffer from symptomatic disease . Current treatments for these “neglected tropical diseases” ( NTD’s ) typically involve periodic mass drug administration ( MDA ) , with the goal of reducing disease prevalence and ideally triggering elimination . Populations in Onchocerca-endemic regions are administered an annual or semi-annual dose of ivermectin; LF-endemic communities normally receive a combination of albendazole with ivermectin in sub-Saharan Africa or with diethylcarbamazine elsewhere [3] . More recently , the World Health Organization ( WHO ) has explored the use of triple-therapy employing ivermectin , diethylcarbamazine and albendazole , recommending its use in specific settings [4] . Recently-approved moxidectin [5] may offer some advantage as a replacement for ivermectin with a more sustained response . These agents primarily kill first-stage larvae ( microfilariae , mf ) and temporarily sterilize adult worms , but do not clear the primary infection . Consequently , MDA must be repeated at regular intervals to successfully affect disease prevalence . Agents that effectively kill adult worms could greatly speed efforts toward elimination of these diseases , and thus are a critical priority for new filariasis drug development . It would also be beneficial to replace ivermectin in significant portions of West Africa that are co-endemic for onchocerciasis and a third filarial disease , loiasis ( caused by infection with Loa loa worms ) . Loiasis typically creates a high burden of circulating mf; treatment of co-infected individuals with ivermectin carries the risk of severe adverse effects or death [6] . An agent that selectively targets adult worms without acutely affecting mf would transform the treatment of these debilitating diseases . Filarial worms causing onchocerciasis and LF carry an obligate symbiotic bacterium , Wolbachia , which is essential for worm fertility and ultimate survival . Clinical studies have demonstrated effective treatment of these diseases through depletion of Wolbachia by anti-bacterial therapy with doxycycline [7 , 8] . Of note , this mechanism has three distinct elements that are considered particularly desirable for a new anti-filarial agent: Since the pathologies of both diseases have been associated with Wolbachia release , an agent that acts by reducing Wolbachia populations within the adult worm may have additional immunological benefits over agents that are directly macrofilaricidal . A slow-kill mode of action reduces the probability of adverse reactions related to sudden worm death . The concept of targeting Wolbachia as an approach to treating filarial disease has been clinically validated using the tetracycline antibiotics doxycycline and minocycline [8]; however these drugs are not ideal for use in the field , as they are contraindicated in children and in women of child-bearing age . The long-term goal of the Anti-Wolbachia ( A·WOL ) Consortium is the discovery and development of novel anti-Wolbachia agents with superior profiles . Recently we reported the discovery of a new anti-Wolbachia compound , ABBV-4083 , derived from the macrolide antibiotic Tylosin A [10] . ABBV-4083 exceeds the efficacy of doxycycline and meets many of the stated pre-clinical goals for a next-generation anti-filarial agent . Herein we describe the details of the discovery program leading to the identification of this novel anti-filarial agent .
Animal experiments using Litomosoides sigmodontis were performed at the Institute for Medical Microbiology , Immunology and Parasitology of the University Hospital Bonn , Bonn , Germany , in accordance to the European Union animal welfare guidelines ( Directive 2010/63/EU and the Amsterdam Treaty: Protocol on the protection and welfare of animals N°33 ) and all protocols were approved by the Landesamt für Natur , Umwelt und Verbraucherschutz , Cologne , Germany ( AZ 84–02 . 04 . 2015 . A507; 84–02 . 04 . 2012 . A140 ) . All pharmacokinetic studies were reviewed and approved by AbbVie's Lake County Institutional Animal Care and Use Committee . Animal studies were conducted in an AAALAC accredited program and veterinary care was available to ensure appropriate animal care . Derivatives of Tylosin A were prepared from TylA ( CAS 1401-69-0 ) or its L- ( + ) -tartrate salt ( CAS 74610-55-2 ) using simple modifications of previously reported procedures , as illustrated in Fig 1 . Selective acylation of the 2’-alcohol ( on the mycaminose sugar ) is accomplished under mild conditions employing acid anhydrides as reagents [11] , presumably as a consequence of neighboring-group activation from the adjacent dimethylamino group [12] . Reaction with dibutyltin oxide forms a cyclic tin oxide between the vicinal diol pair at 3” and 4” ( mycarose sugar ) ; this serves as an activated intermediate for the selective acylation of the 4”-hydroxyl group [13] . Alkylation of this site is also possible , under more forcing conditions and using active alkylating agents . When the 2’-substituent is acetyl , a free 2’-hydroxyl group may be liberated via heating in methanol . This transformation is accelerated through the addition of a small amount of solid NaHCO3 . This straightforward sequence of transformations allows for the preparation of 2’- , 4”- , or 2’/4”-modified tylosin analogs , from common intermediates , in good overall yields ( Table 1 ) . Compounds were screened for anti-Wolbachia activity in vitro in the A·WOL-validated Wolbachia-infected Aedes albopictus ( C6/36 wAlbB ) 7-day cell-based assay which utilizes a 384-well format assay with high content imaging ( HCI ) ( Operetta ) as described previously [14] . PO doses were administrated by oral gavage , IP doses by intraperitoneal injection to BALB/c mice or Sprague-Dawley rats ( Charles River Laboratories , USA ) . Serial blood samples collected into EDTA anticoagulant for plasma concentration analysis were obtained from each animal after dosing . EDTA preserved plasma samples were extracted by protein precipitation with acetonitrile fortified with internal standards . The supernatant was injected into an HPLC-MS/MS system for separation and quantitation . Detection was accomplished using a triple quadrupole mass spectrometer operated either in electrospray or atmospheric chemical ionization ( APCI ) mode . The area under the plasma concentration-time curve ( AUC ) was calculated using the linear trapezoidal rule . Mice and jirds ( Meriones unguiculatus , both obtained from Janvier , Saint-Berthevin , France ) were kept in individually ventilated cages with food and water ad libitum and a light/dark cycle of 12h . As described previously [15] , female BALB/c mice or female jirds were infected at 6–8 weeks of age with L . sigmodontis larvae through the bites of Ornithonyssus bacoti mites . The same batch of L3 larvae-containing mites were used for all experimental groups within each experiment to ensure comparable rates of infection . One day post infection , mice were dosed IP or PO with TylA ( Sigma Aldrich , 200 mg/kg BID x 7 days ) , doxycycline hyclate ( Sigma-Aldrich , 200 mg/kg BID x 14 days ) , or vehicle using a volume of 10 ml/kg . At 35 days post-infection mice were euthanized using an overdose of isoflurane; worms were recovered from the pleural cavity by pleural lavage , counted , sexed and staged for development into L4 and adult worms based on the difference of the buccal capsule through microscopic examination . Female worms were measured for length ( mm ) as a marker for development as previously described [16] . Data were distributed in a non-parametric fashion , median and interquartile ranges are presented . For comparing the length of the female worms , the Mann-Whitney-U test was used to calculate statistical differences either against the vehicle treated or gold standard groups . Starting at 14 weeks post infection , microfilariae-positive jirds ( n = 7 per group ) were dosed PO with ABBV-4083 ( 150 mg/kg QD ) dissolved in 0 . 5% HPMC/0 . 02% Tween-80 or vehicle using a volume of 5ml/kg . Microfilariae numbers were assessed through visual inspection of blood samples collected from the saphenous vein at weekly intervals post-dosing . For this , 10 μl of peripheral blood were diluted in 300 μl of Hinkelmann solution ( 0 . 5% Eosin Y , 0 . 5% Phenol , 0 . 185% Formaldehyde in aqua dest ) . After 5 minutes of centrifugation at 400g , the supernatant was discarded and the pellet transferred for microscopic quantification of the microfilariae . At 16 weeks post-treatment , jirds were euthanized by an overdose of isoflurane; worms were recovered from the pleural cavity and counted . Remaining intact female adult worms were used to assess embryogenesis and the Wolbachia load . For the latter , genomic DNA ( gDNA ) was extracted from individual female adult worms and quantification of the Wolbachia ftsZ ( wLs-ftsZ ) and L . sigmodontis β-actin ( Ls-act ) gene copy numbers was performed by quantitative real-time PCR ( qPCR ) [16] . For embryograms , remaining intact female adult worms were individually homogenized in 20% Hinkelmann/80% PBS solution , diluted 1:10 in PBS and quantified by microscopy . Embryonal stages were differentiated as egg , morulae , pretzel and stretched microfilariae [17] .
We began our work by selecting a diverse and representative sample of the AbbVie antibiotics collection ( 129 compounds ) for single-point testing against Wolbachia pipientis in an insect cell line [14] . This screen revealed several novel leads , most notably the established veterinary antibiotic Tylosin A ( 1 , Fig 1 ) . While TylA has a long history of use in multiple animal species , it has never been studied in humans; and its activity against Wolbachia has not previously been reported . It is a potent anti-Wolbachia agent , with an EC50 value of 28 nM ( measured in Wolbachia-infected insect cells as described above ) , similar to that for doxycycline . Other commercially available 16-membered macrolides ( spiramycin , josmycin , midecamycin and leucomycin ) are inactive against Wolbachia; similarly 58 semisynthetic leucomycin derivatives from the AbbVie collection showed no activity at 10 μM concentration . Notably , none of these macrolides contain the mycinose sugar present in TylA . In contrast , Tylosin B ( TylB ) , which contains the mycinose residue but lacks the mycarose sugar of TylA , retained substantial though reduced activity against Wolbachia in vitro ( EC50 88 nM ) . As follow-up to this initial in vitro study , we examined the activity of TylA in a mouse model of filarial disease [15] . Mice naturally-infected with L . sigmodontis through mite bites were treated with TylA or doxycycline at a dose of 200 mg/kg twice daily ( Fig 2A ) . When TylA was dosed IP for 7 days , recovered worms were notably shorter than controls , indicating that development has been suppressed . This result is similar in magnitude to that produced by 14 days of doxycycline treatment ( Fig 2B ) . Oral dosing of TylA , however , produced a minimal response . Supplementary experiments have correlated this growth stunting phenotype with a reduction in Wolbachia levels [10] . These results were readily explained by examination of circulating drug levels measured in a companion pharmacokinetic study ( Fig 2C ) . Drug levels in the IP arm of this study were >30-fold higher than those achieved when the drug is given PO . We suspect that the poor oral bioavailability of TylA results from an inability to efficiently cross membranes like the gut lining; the compound exhibits very low permeabililty ( <0 . 1X10-6 cm/sec ) in a canine kidney cell monolayer system ( MDR-MDCK ) . Therefore , improving drug absorption by increasing permeability became a primary goal for our lead-optimization studies . We focused our initial Structure-activity relationship ( SAR ) studies on modifications that reduce the H-bond donor capacity of our lead; hypothesizing that the large number of free hydroxyl groups ( TylA has 5 free–OH’s ) was responsible for the poor permeability ( and thus poor bioavailability ) of TylA . The most readily accessible of these hydroxyl groups is the 2’-OH ( on the mycaminose sugar ) , which is internally activated by an adjacent amine functionality . Thus , as previously reported by Tsuchiya and others , this position may be acylated under mild conditions [11 , 12] . Acylation of the 2’-position causes a modest but significant loss of potency against Wolbachia; esters 2a-2c ( Table 1 ) have EC50 values that are 2–3 fold higher than parent TylA ( Table 2 ) . To test our hypothesis regarding the role of the free hydroxyl groups in impeding the uptake of TylA , we compared the pharmacokinetic profiles of 2c and TylA in rodents ( Table 2 ) . Acylation of the 2’-position leads to a 6-fold improvement in plasma drug levels ( as expressed by total area-under-the-drug-exposure-curve , AUC ) . In the study of compound 2c we also looked for the presence of TylA ( the de-acylated metabolite ) , determining that an ester group at this position is relatively metabolically stable . This improvement in drug exposure is enough to override the modest loss of potency that comes with acylation; the potency-weighted AUC ( determined as AUC/EC50 ) is ~2 . 6 times higher with the 2’-valerate ester , and the potency-weighted 8-hr drug level ( determined as C8hr/dose/EC50 ) doubles . We have previously noted that maintaining free drug levels above the EC50 value is an important determinant of in vivo efficacy [10] . This early result seems to support our central hypothesis , and encouraged us to explore the effect of modifying other hydroxyl groups in TylA . We had notable success through modification of the 4”-OH , on the mycarose sugar . Once the 2’-OH has been derivatized , selective activation of the 4”-position is possible through formation and acylation of a 3”/4”-cyclic tin complex ( e . g . compound 3a/b , Fig 1 ) , as previously described by Kiyoshima et al . [13] When the 2’-substituent is an acetyl group , the corresponding 2’/4”-diacylated analog 4a ( Table 1 ) may be selectively deacylated at the 2’-position simply by warming in methanol , to give the 4”-mono-ester 5a . Unexpectedly , modification of this 4”-site significantly improved the activity of the resultant derivatives against Wolbachia; for example , ester 5a has an in vitro EC50 of 1 . 3 nM ( Table 2 ) , 25-fold lower than TylA . The pharmacokinetic profile of compound 5a was examined in mice . As with the previous study of 2’-ester 2c , we observed that oral drug levels increase ( ~3-fold ) upon 4”-acylation ( Table 2 ) . However , in this case the primary drug measured in the plasma is not the parent ester; rather , it is the deacylated metabolite 1 ( TylA ) . In fact , no sign of parent is observed at any time point in this study , suggesting a very rapid cleavage of the 4”-ester moiety . The metabolic susceptibility of these 4”-esters ( presumably to hepatic esterases , though this has not been proven ) is substantially higher than that of the corresponding 2’-esters , though the latter are more susceptible to chemical hydrolysis . While it is possible that this metabolic pathway is rodent-specific , the result suggested to us that another solution was desirable . To this end , we explored several strategies for modifying the 4”-position with substituents expected to have greater metabolic stability . The following representative experiment is illustrative of the in vivo activity of ABBV-4083 . Litomosoides sigmodontis is a filarial parasite that leads to patent infections in BALB/c mice and gerbils ( jirds ) ; natural infection can be established through mites [15] . When infected jirds ( Meriones unguiculatus ) are treated with ABBV-4083 at an oral dose of 150 mg/kg , once daily for 14 days , Wolbachia levels ( measured 16 weeks post-treatment-initiation , pti ) are reduced by >99 . 9% in the recovered female adult worms ( Fig 3A ) . As predicted , these reductions in symbiont levels had consequences for worm fertility . Starting at ~7 weeks pti , levels of circulating microfilariae declined ( Fig 3B ) and were completely cleared from 12 weeks pti until the end of this study at 16 weeks pti . Control animals maintained circulating levels of microfilariae throughout the study . We have previously demonstrated that ABBV-4083 is not directly microfilaricidal [10] , so it is likely that this decrease is a consequence of a loss of worm fertility . In fact , analysis of the uterine contents of female worms ( “embryograms” ) indicate a profound effect on embryogenesis ( Fig 3C ) , as suggested by the near-complete loss of all embryonic forms including eggs . Additional experiments [10] demonstrate that ABBV-4083 equals or exceeds the efficacy of doxycycline with regard to Wolbachia depletion and maintenance of microfilariae clearance even when the latter is dosed for substantially longer intervals ( e . g . 14- vs 7-days ) , strongly suggesting the possibility that ABBV-4083 might provide a shorter-course treatment for filarial diseases . As a preclinical candidate , ABBV-4083 has been evaluated in a variety of in vitro assays assessing preclinical safety . In an initial battery of 35 assays assessing the general selectivity of the compound , there were no significant interactions with any receptors at a maximum concentration of 10 μM . This pattern was confirmed in studies across 77 mammalian receptors , ion channels , enzymes and transporter assays , in which a significant interaction was only observed in two assays [10] . ABBV-4083 did not inhibit functional hERG channel activity at a maximum concentration of 30 μM , and did not produce significant cardiovascular effects when administered to dogs . The compound was neither mutagenic nor clastogenic in in vitro genotoxicity screening assays . No potential to induce phospholipidosis was observed in vitro , and the compound did not induce steatosis in an in vitro high-content screen . In preparation for first-in-human studies , the safety of ABBV-4083 has been extensively evaluated in 28-day GLP general and reproductive toxicity studies . In addition , the synthesis has been adapted to produce GMP quality supplies . Through properties-driven optimization of the anti-Wolbachia lead Tylosin A , we have identified ABBV-4083 , an analog with a superior pharmacokinetic profile and remarkably improved potency . This combination of improved properties addresses the liabilities of TylA itself , and the analog appears suitable for use as an oral therapeutic for treating onchocerciasis and/or lymphatic filariasis . Based on preclinical data , ABBV-4083 demonstrates potential improvements over the use of doxycycline as an anti-Wolbachia agent in terms of both safety and reduced treatment duration . Given the short synthesis of this compound from a widely available and inexpensive veterinary product , its use for neglected diseases such as onchocerciasis and lymphatic filariasis should not be limited by cost of goods . Whether ABBV-4083 is best suited for MDA or test-and-treat strategies will only become evident after clinical trials defining its efficacy and safety . Phase 1 studies of this agent in normal healthy human volunteers are currently underway; results will be reported in due course .
|
The Wolbachia bacterium lives symbiotically within the filarial worms that cause onchocerciasis and lymphatic filariasis . In the absence of these bacteria juvenile worms cannot mature , females are unable to reproduce , and the worm life-span is significantly shortened . Thus , anti-Wolbachia therapy would seem to be an ideal approach to treating filarial disease . This concept has been validated clinically using the tetracycline antibiotic doxycycline . However , doxycycline , which is contraindicated in children and women of child-bearing age , is not ideal for field use . Additionally doxycycline requires a long course of treatment ( minimum 4 weeks of daily use ) to provide clinical benefit . A safer , faster anti-Wolbachia agent would be a valuable addition to the filariasis pharmacopeia . Through targeted screening , we have identified the veterinary antibiotic Tylosin A ( TylA ) as an effective anti-Wolbachia lead compound . While the in vitro and in vivo activity of TylA match that of doxycycline , its potency is not ideal , and it suffers from poor oral bioavailability . By appropriate derivatization of the 4”-OH group of TylA ( on the mycaminose sugar ) we were able to improve oral absorption while simultaneously increasing anti-Wolbachia potency . Optimization of this substituent with a focus on metabolic stability led to the identification of ABBV-4083 , the 4”- ( 4F-benzyl ) analog . ABBV-4083 is exquisitely active against Wolbachia , with an improved pharmacokinetic profile . This analog outperforms doxycycline in several animal models of filariasis; in particular it clears the bacteria effectively with substantially shortened dosing regimens . Preclinical efficacy and safety studies indicate that ABBV-4083 may have a desirable profile as a novel , next-generation anti-filarial agent .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
] |
[
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"drugs",
"tropical",
"diseases",
"microbiology",
"parasitic",
"diseases",
"acylation",
"wolbachia",
"antimalarials",
"esters",
"antibiotics",
"neglected",
"tropical",
"diseases",
"pharmacology",
"onchocerciasis",
"bacteria",
"drug",
"metabolism",
"veterinary",
"science",
"doxycycline",
"veterinary",
"diseases",
"proteins",
"chemistry",
"pharmacokinetics",
"biochemistry",
"helminth",
"infections",
"post-translational",
"modification",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] |
2019
|
Discovery of ABBV-4083, a novel analog of Tylosin A that has potent anti-Wolbachia and anti-filarial activity
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.